Studies were done in dryland farming to represent agricultural practices commonly done by small-scale farmers in South Africa, who cannot afford to establish irrigation systems. The experiments were set on a land of 17 × 15.5 m. The land was ploughed and disked a month earlier. Plots 3 m wide and 4 m long, giving an area of 12 m², were made using hand hoes. Organic fertilizers (co-compost and chicken manure) were applied based on crop N requirements recommended by the Fertility and Advisory Services (FAS) of the KwaZulu-Natal Department of Agriculture and Rural Development (DARD) soil analysis results. Organic fertilizers have a slow nutrient release rate, so the application rate was calculated following Equation 1 according to the methods recommended by Tesfamariam et al. (2020).

In Equation 1, the mineralization rate for the BALL site (29%) was obtained from Ogbazghi et al. (2016) with an assumption that Pietermaritzburg is within the sub-humid region. The compost N content was obtained from the FAS analytical results. The organic fertilizers were applied on each planting station at a rate of 20 tons per hectare a month prior to planting.

Chilies (Capsicum annuum) of the variety Star 6,604 was selected as a test crop due to its potential for processing into chili paste and powder by the farmers at BALL, and according to the WHO (2006) multibarrier approach specifications have lower microbial contamination risks. The seedlings obtained from a local registered nursery were planted on 17 November 2022 at a spacing of 0.5 m × 0.5 m, and the test crop lasted for 6 months. The inorganic fertilizers were applied at recommended rates (Urea: 326 kg ha⁻¹, Single Super Phosphate (SSP): 1,960 kg ha⁻¹, Potassium Chloride (KCl): 558 kg ha⁻¹), which directly translates to 150:196:290 (N: P: K). The urea was split-applied; 50% was applied a week after planting, the other 50% was applied a month after planting (25%) and 3 months after planting (25%). The SSP and KCl were applied once off at planting. The fertilizers were banded and covered up by the soil soon after application. The urine was stored for 1 month at room temperature, as recommended by WHO (2006). The stored urine N content value of 3.9 g L⁻¹, which was obtained from previous studies (Odindo et al., 2022) was used to calculate crop N requirements following Equation 2.

The urine was diluted with water at a ratio of 3:1 and applied at the same time as inorganic fertilizers (a week, a month, and 3 months after planting), directly to the soil surface of each plant using the watering container to avoid foliar contact (Richert et al., 2010). An integrated weed, pest, and disease management program was implemented. The selected chili variety has a general tolerance to leaf diseases; however, agrochemicals were not used to control pests and diseases because they were not identified.

The data obtained from the National Aeronautics and Space Administration (NASA) access viewer database (NASA, 2022) was used to monitor climatic parameters for optimum crop growth. Comparisons were made between the amounts of rainfall received on the site during the crop growing period with historic monthly mean rainfalls averaged for >30 years. Water balances were done by calculating the monthly average crop evapotranspiration following Equation 3.

Whereby the ET_crop is the historic average crop evapotranspiration, ET_pan is the historic average reference evapotranspiration measured from the class A evaporation pan, and K_c is the crop factor value for the chili crop. A simple water balance assessed the rainfall sufficiency for crop production by subtracting average monthly ET_crop from actual monthly rainfall.

Soil chemical and biological properties were measured before planting and after harvesting. Samples for chemical analyses were collected within the 0.3 m soil depth from five different spots within each plot. These were bulked to form composite samples for each experimental unit. The samples were submitted to the Fertility and Advisory Services (FAS) of the Department of Agriculture Land Reform and Rural Development laboratory for the analysis of organic carbon, cation exchange capacity, nutrients (organic N, mineral N, extractable P, extractable K, Ca, Mg, Cu, Al and Fe), soil pH (KCl), acid and base saturation according to the standard methods (Manson et al., 2020).

The chili harvesting started 12 weeks after transplanting. Only red chilies were harvested, leaving green ones and flowers. The fresh mass from the harvested chilies was measured immediately using a balance with ±0.02 g accuracy. The yield per harvest was determined according to Equation 4.

Whereby the “M” is the mass of chilies in grams per plant. “A” is the area per plant, referring to plant spacing of 0.5 × 0.5 m (0.025 m²). The cumulative yield was then determined as the sum of mean yield values from all the treatments and three harvests obtained from the tables of means calculated by the GenStat software.

About 200 g of soil was sampled from the 0.3 m topsoil level, within a 0.1 m radius of the plant, before planting and after the final crop harvest. Each sample was placed in zip-lock plastic bags at room temperature for not more than 5 days and submitted to Sporatec Soil Microbiology Laboratories within the Department of Microbiology at Stellenbosch University to analyze microbial communities and soil enzymatic activity. Microbial communities are important in nutrient recycling. The more even the taxa, the more stable it is to continually carry ecosystem services (Drosos et al., 2023). Therefore, microbial communities were analyzed using the molecular fingerprinting technique, which generated a profile for each composite soil sample (Singh et al., 2006).

Soil biochemistry is an important indicator of soil health that can help understand the influence of certain soil amendments on soil ecology, which drives optimal microbial processes. When organic matter decomposes in the soil, microorganisms release various enzymes that break down complex molecules such as urea, cellulose, and phosphates, producing carbon dioxide and inorganic nutrients (Yu et al., 2025). Therefore, N mineralization, urease activity, β-glucosidase, PO₄³⁻-P, alkaline and acid phosphatase, microbial activity, and organic and active carbon were measured according to standard methods (Karlen et al., 2021). The aerobic incubation method was used to determine soil mineralization after incubating soil samples at 25°C for 14 to 28 days. Initial and final concentrations of inorganic nitrogen (NO₃⁻-N and NH₄⁺-N) were extracted with 2 M KCl and measured colorimetrically using the Thermo Scientific™ Gallery™ Discrete Analyzer, to calculate net N mineralization (Maynard et al., 2007). The β-glucosidase was determined by incubating a soil sample with a substrate containing p-nitrophenyl-β-D-glucopyranoside in a buffer solution at pH 6, 37°C and for 1 h. The reaction was stopped by adding calcium chloride, and the absorbance was measured at 400 nm using a spectrophotometer. The Bray II method was used to test the PO₄³⁻-P, after shaking a soil sample with a solution of 0.03 M NH₄F and 0.1 M HCl and then filtering. The Murphy-Riley method was used to colorimetrically measure PO₄³⁻-P at 882 nm absorbance. Alkaline and acid phosphatases activity was determined after incubation in a buffer solution, which was acidic (acid phosphatase) or alkaline (alkaline phosphatase), and p-nitrophenyl phosphate as the substrate. The samples were measured spectrophotometrically at 405 nm to quantify the enzyme activity. The fluorescein diacetate (FDA) hydrolysis assay was used to estimate total microbial activity in soil by measuring the enzymatic conversion of FDA into fluorescein. The amount of fluorescein produced, quantified spectrophotometrically, reflects the collective activity of microbial enzymes such as esterases, lipases, and proteases (Adam and Duncan, 2001). The Walkley-Black method was used to measure organic C, and the potassium permanganate (KMnO₄) oxidation was used to measure active and labile C forms.

Groundwater samples were analyzed for suitability as irrigation water and human health safety by the drinking water quality standards. The borehole water samples were collected before planting and after 6 months. These were submitted to the salinity laboratory of the FAS for the analysis of salinity, Na, Ca, Mg, K, Cu, Mn, Fe, Zn, Total alkalinity, SAR, salinity class, and Cl⁻ according to standard methods (APHA AWWA & WEF, 2017). Using human excreta products is likely to contaminate groundwater with pathogens. Water samples were monitored at a six-month interval and analyzed for nutrients (N and P) and pathogens (E. coli) according to Velkushanova et al. (2021).

The mean differences in organic C are reported in Table 5. Comparisons amongst the treatments showed significant differences (p < 0.05) in organic C between no fertilizer and co-compost + urine (Table 5). The soil organic C significantly increased with the application of co-compost + urine compared to no fertilizer, urine, or fertilizer treatments. This is because co-compost contains high carbon content (Table 2). Depending on the forms of organic C, the compost plays a role in sequestrating C when non-digestible forms are left to accumulate in the soil (Zhao et al., 2020). In this case, the active organic C fractions did not significantly differ across all treatments (Table 5). The stable organic C forms increase water and nutrient retention capacity, which is important for climate change adaptation.

The microbial activity was significantly lower in co-compost + urine treatment compared to chicken manure amendment, despite the two differing insignificantly in the observed soil organic C value (Table 5). This observation is frequently reported, as microbial activity is regulated by a combination of factors, including the physical, chemical, and biological properties of the soil (edaphic factors) and the quality of the substrate (Wang et al., 2024; Furtak and Gałązka, 2019; Muneer et al., 2022; Cao et al., 2016). The study was done on one soil type, ruling out the effect of soil properties as an explanation. Soil moisture content also plays an important role in microbial activity. Amending soils with organic matter increases soil moisture retention capacity, further stimulating microbial activity in a water-limited soil (Martín-Lammerding et al., 2021; Hueso et al., 2012). All the treatments were subjected to dryland conditions, but the soil moisture content did not significantly differ (p > 0.05) across all treatments (Table 5). Even the microbial activity in the no-fertilizer treatment did not significantly differ from the co-compost + urine (Table A1). These findings explain that, while it is recognized that organic amendments increase organic matter, which boosts microbial activity, their impact within one season was not effective. Even most studies that looked at the effects of organic amendments on microbial activity were done over long term of at least 2 years (Desjardins et al., 2025, Sayre et al., 2023, Wang and Jiao, 2022). Further, calling for long-term monitoring of the relationship between organic fertilizer amendment and improved soil moisture retention in dryland agriculture.

Variations in organic fertilizer quality, such as C: N ratio, pH, nutrient content, organic matter decomposability, type of organic matter, and maturity, contribute to microbial activity. Characteristics of organic amendments used are shown in Table 2. Chicken manure has high N content (5%) compared to co-compost (1.2%). The C: N ratio plays a role in microbial activity; soil amendments with high C: N ratios have low decomposition rates because microbes have to source N from somewhere, while low C: N ratios provide a balance between required C and N (Li et al., 2018), and initiate protein synthesis required for enzymes and reproduction (Wang and Jiao, 2022), leading to higher microbial biomass. The higher NPK content in chicken manure (Table 2) explains higher microbial activities in the soil compared to the use of co-compost, regardless of the time frame. When it comes to the time frame Jin et al. (2022), found that a single-season application of chicken manure was enough to change the microbial soil processes.

Another possibility could be the stability of the amendment used. Poorly decomposed organic amendments with high moisture content trigger microbial activities because of high concentrations of readily available organic matter, nutrients, microbial loads, and a shift in soil microbial structure (Liu X. et al., 2023). The co-compost used was stabilized by composting for over 5 months, which allowed organic matter to degrade; hence, its application in soil could not trigger microbial activity, just like unstabilized chicken manure. Urra et al. (2019) confirmed that fresh unstabilized chicken and horse manure increase soil microbial activity due to the presence of organic N and P forms. Some studies attributed higher microbial activity in chicken manure-amended soil to changes in soil microbial community structure (Chen et al., 2023; Jin et al., 2022). The studies that reported similar results of chicken manure, although it was not clearly mentioned to be composted, were done for 2 years in soils with a pH of 8.8 (Chen et al., 2023).

The presence of xenobiotics such as heavy metals and micropollutants sewage sludge-derived co-compost is raising significant concerns about their ecotoxicity. High concentrations of heavy metals affect soil microbial activities by disrupting cationic exchange processes while altering soil pH (Abd-Elhalim et al., 2025). However, the sewage sludge-derived co-compost was low in heavy metals, which was below the South African fertilizer legislation standard limits (Table 2). Even the co-compost pH was within the optimum ranges for agricultural use, and the reported soil pH did not significantly differ (Table 5). Meaning that the co-compost was environmentally safe for agricultural use. However, Panneerselvam et al. (2022) discussed the importance of micropollutants such as drugs and pesticides, which might find their way into the wastewater treatment system, contaminating the resulting sludge. Micropollutants are deemed emerging contaminants because there are no consistent regulations to guide their use against ecotoxicity (Yu et al., 2025; Häfner et al., 2023; Gul et al., 2022). This is one of the grey areas in the use of human excreta-derived fertilizers, requiring further studies.

Higher microbial activity is not always good for agriculture because increased microbial biomass immobilizes inorganic N, making it unavailable for crop uptake. Meaning that co-compost is a good soil amendment in dryland agriculture because of its C: N ratio from mixing organic garden waste and sewage sludge.

Closing nutrient loops and enhancing soil health using biobased fertilizers such as urine and co-compost is important. The findings showed that even the microbial activity in co-compost + urine, urine, and fertilizer treatments was significantly (p < 0.05) lower than no-fertilizer amendment (Table 5). This was not expected because organic carbon from co-compost and mineral nutrients provided by urine were supposed to boost microbial activity. There are mixed findings on this matter in the literature. Yu et al. (2025) found that the application of human urine in soil can increase microbial activity due to the presence of urea, which stimulates urease activity, as well as the presence of dissolved organic carbon and a shift in soil pH due to an increase in electrical conductivity. This was reported in similar studies, whereby the addition of bovine urine increased microbial activity after stimulating ammonium-nitrifying bacteria (Orwin et al., 2010). Despite high soil salinity due to the use of urine, the microbial activity still increases in a similar way to applying inorganic fertilizers (Rumeau et al., 2024), which is consistent with the lack of significant differences between urine and fertilizer treatments. However, higher microbial activity in non-amended soil shows that the application of amendments altered natural processes in the short run. A meta-analysis by Liu W. et al. (2023) found that there are various factors that can stimulate microbial activity, an initial change in soil pH being one of them. However, this cannot be a cause in this case because the soil pH did not significantly differ across all treatments (Table 5). As confirmed in literature, sometimes the application of urine fertilizers elevates the soil salinity, shifting the microbial balance by reducing the microbial balance of sensitive microbes such as Bacillota (Yu et al., 2025). The same applies to the use of inorganic fertilizers (Carrascosa et al., 2023; Lian et al., 2022). A > 25% decline in microbial activities in the soil amended with inorganic fertilizers was also reported (Elbl et al., 2019). A stable soil ecosystem is created when inorganic fertilizers are applied together with manure or organic amendments (Lian et al., 2022). Reduction of microbial activity after the use of co-compost and urine could have been triggered by microbial stress. Microbial stress in this context refers to the ability of microorganisms to adapt to adverse conditions such as pH, temperature, and other extreme conditions (Abdul Rahman et al., 2021). Some microorganisms may succumb due to their inability to synthesize proteins that help them adapt to environmental stresses, as was evidenced in studies done using cow urine (Bertram, 2009). However, in this juncture, the influence of soil salinity in response to the use of urine is not a factor because the electrical conductivity did not significantly differ (p > 0.05) across all treatments (Table 5). Underscoring the need for further investigations into this matter with longer-term studies, as proposed by Yu et al. (2025).

The sum of metabolic processes taking place in the soil can be determined by enzymatic activities. This section digs deep into specific enzymes responsible for nutrient cycling. During microbial degradation of organic matter in the soil, enzymes such as β-Glucosidase are released to break down β-1,4-glycosidic bonds in glucosides, oligosaccharides, and various glycoconjugates, including cellulose from organic matter (Adetunji et al., 2017; Bastida et al., 2008; Ansari et al., 2023). When it comes to nitrogen cycling, urease enzymes hydrolyze urea into ammonium-N, which, depending on soil ecological factors such as pH, can further be nitrified by nitrifying bacteria into plant bioavailable forms (Yu et al., 2025). However, both the urease and the β-Glucosidase activities in this study did not significantly differ across all the treatment groups (p > 0.05) (Table 5). This further explains that the higher microbial activities in chicken manure were not linked to the amendments applied. This is further supported by the fact that the soil active carbon (labile carbon) pool, which serves as the readily available food source for microbes (Orwin et al., 2010; Sayre et al., 2023; Li et al., 2024) did not significantly change in response to the amendments applied. This implies that the application time frame and loading rates could have been very short to show significant changes in the soil. Most studies were done for 2 years or beyond (Ansari et al., 2023; Liu W. et al., 2023; Drosos et al., 2023). Therefore, the six-month experimental period, although during the rainy season of the study site, which received 1,656 mm of rainfall (Figure 1C), was not enough to influence soil health.

Table 5 reports on N mineralization and extractable P results. The N mineralization rate did not significantly differ amongst all treatments (p > 0.05), while extractable P significantly differed (p < 0.05). Chicken manure and co-compost had significantly higher median values of extractable P. This is attributed to relatively high extractable P concentrations in organic amendments (Table 2). This cannot be linked to P mineralization, because the acid and alkaline phosphatase activities did not significantly differ in all treatments (Table 5). Several studies have confirmed that the P in untreated urine is generally lower than N and K, and these are lower than in an inorganic fertilizer (Jönsson et al., 2004; Richert et al., 2010; Alemayehu et al., 2020). Low concentrations of extractable P in soils that were applied inorganic fertilizer resulted from high solubility of Single Super Phosphate, uptake by plants, or loss from the topsoil. Organic matter increases the surface area for soil P adsorption capacity, allowing soils to capture and retain inorganic phosphates. This explains why the co-compost and urine combination and chicken manure had higher extractable P median values. The ability of organic amendments such as co-compost to improve nutrient retention by creating adsorption sites that retain cations and anions is well-documented and recommended from environmental and agronomic perspectives (Urra et al., 2019; Snyman et al., 2006; Fuhrmann et al., 2022). The only challenge is when phosphorus is transported to nearby water bodies through surface runoff, and causes pollution if proper slope management strategies are not put in place (Sharpley, 2016).

Microbial activity may not be a good indicator of soil health in the short run, as it did not show the activity of beneficial microbes around nutrient cycling. Lower microbial activities were observed in co-compost, urine, and even in inorganic fertilizer-amended soils, while relatively higher in the chicken manure and even in unamended soils. Some other soil biodiversity indicators, such as the Shannon Index, the Simpson Index, and Species richness, can be used as soil health indicators. The Shannon index quantifies the diversity of microbial communities by integrating the number of different species (richness) and their relative abundance (Meng et al., 2016). This has been used in various agroecosystems to assess the effects of various management practices on soil microbial diversity (Meng et al., 2016; Sayre et al., 2023; Drosos et al., 2023). These indicators did not significantly differ across all the treatments (Table 5), indicating that the short-term use of the organic amendments did not significantly change microbial diversity, regardless of notable differences in soil organic carbon compositions and microbial activity. This clearly shows that the microbial responses to organic amendments was not significant over the studied period.