Field experiments were conducted during the 2019–2021 growing seasons in the Piskent district of the Tashkent region, Uzbekistan (40°58′ N, 69°21′ E) (Figure 1). The study area is located in the piedmont zone and is characterized by irrigated agricultural lands with varying slope gradients, making it highly susceptible to irrigation-induced erosion (Isaev et al., 2020; Mirzayev, 2018).

The experimental fields were situated on meadow-sierozem soils, which are widely distributed in the foothill regions of Central Asia. These soils are classified as moderately to heavily loamy in texture and are known for their relatively low organic matter content and high erosion vulnerability under improper irrigation management (Mirzayev, 2018). The bulk density of the plough layer (0–30 cm) ranged from 1.32 to 1.35 g cm^-3, while total porosity varied between 48% and 50%.

The climate of the region is sharply continental, with hot, dry summers and precipitation concentrated mainly in winter and early spring. During the vegetation period (April–October), the mean air temperature ranges from 24 to 26 °C, and the average annual precipitation is approximately 350–400 mm (Isaev et al., 2020).

The experiment was arranged as a two-factor factorial design using a randomized complete block design (RCBD). The treatments consisted of the following factors: Factor A (slope gradient): (i₁) 1.5°, (i₂) 2.5°, and (i₃) 3.5°. Factor B (irrigation method): (i) Conventional Straight Furrow Irrigation (CSFI), used as the control; (ii) Zig-Zag Furrow Irrigation (ZZFI), representing the experimental treatment.

In the ZZFI treatment, the furrows were initially cut using a standard KKhU-4 cultivator mounted on a tractor to establish the primary furrow lines. Subsequently, the specific zigzag geometry was shaped manually using hoes to create artificial bends at intervals of 5–6 m along the furrow length. This manual adjustment ensured the formation of a consistent serpentine flow path to increase hydraulic resistance, following the approach proposed by Sepaskhah and Kamgar-Haghighi (1997).

Each treatment combination was replicated three times. Individual plot size was 0.10 ha, and the total experimental area covered 1.8 ha. Treatments were randomly allocated within each block to minimize spatial variability effects, in accordance with standard field experiment methodology (Dospekhov, 1985).

Cotton (Gossypium hirsutum L., cv. ‘Sulton’), a locally adapted variety with high fiber quality, was used as the test crop. Sowing was carried out using a row spacing of 90 cm with 10 cm intra-row spacing, resulting in a plant density of approximately 80,000–90,000 plants ha^-1.

Mineral fertilizers were applied uniformly across all treatments at recommended rates: 200 kg N ha^-1, 140 kg P₂O₅ ha^-1, and 100 kg K₂O ha^-1 (Table 1). Phosphorus and potassium fertilizers were applied with 70% as a basal dressing prior to sowing and 30% during the flowering stage. Nitrogen was split into three applications: 20% at sowing, 30% at the squaring stage, and 50% during flowering, following regional agronomic recommendations (Isaev et al., 2020).

Irrigation water discharge was measured at the head of the furrows using a Chippoletti trapezoidal weir with a measurement accuracy of ±2%. Soil moisture content was determined before and after each irrigation event using the gravimetric method by drying soil samples at 105 °C to constant weight.

To quantify soil erosion, runoff samples (1 L) were collected from the tail end of the furrows during each irrigation event. The samples were filtered and dried to measure the suspended sediment concentration. Total soil loss was calculated based on runoff volume and sediment concentration.

Statistical analysis was performed using one-way ANOVA to determine the effects of slope and irrigation method on soil erosion, water use, and crop yield. The means were compared using Fisher’s Least Significant Difference (LSD) test at a significance level of p < 0.05. All statistical computations were conducted using the SPSS software package.

Irrigation water discharge was measured at the inlet and outlet of each furrow using Cipolletti trapezoidal weirs with a measurement accuracy of ±2%. In both irrigation treatments (CSFI and ZZFI), water was applied at a consistent inflow rate ranging from 0.10 to 0.20 L s^-1 per furrow to prevent overflow while ensuring water advance to the tail end.

Soil erosion caused by irrigation was quantified by collecting runoff samples at the downstream end of the furrows. During each irrigation event, 1 L water samples were collected at hourly intervals and transported to the laboratory. Suspended sediments were determined by the evaporation and gravimetric method (drying at 105 °C). The total soil loss (M, t ha^-1) was calculated based on runoff discharge, sediment concentration, and irrigation duration, following established procedures described by Trout et al. (1993) and Mirzayev (2018) (Equation 1):

All experimental data were analyzed using a two-way analysis of variance (ANOVA) to evaluate the main effects of slope gradient (Factor A) and irrigation method (Factor B), as well as their interaction. Statistical processing was performed using GenStat and Microsoft Excel software packages.

Differences between treatment means were considered statistically significant at p < 0.05. Post-hoc comparisons were conducted using Fisher’s Least Significant Difference (LSD_0.05) test, in accordance with the methodological guidelines for field experiments by Dospekhov (1985). In the results section, significant differences are indicated by different letters (a, b, c) above the bars in figures and tables.

Field measurements revealed a strong dependence of irrigation flow behavior on both slope gradient and furrow configuration. Under Conventional Straight Furrow Irrigation (CSFI), an increase in slope steepness led to a marked rise in flow velocity along the furrow length. Notably, at the steepest slope gradient of 3.5°, measured flow velocities exceeded the critical non-erosive threshold (V > 0.12 m s^-1), resulting in unstable flow conditions and intensive sediment transport.

In contrast, the Zig-Zag Furrow Irrigation (ZZFI) treatment significantly altered hydraulic conditions across all slope gradients. The introduction of periodic bends along the furrow increased hydraulic resistance (roughness), which effectively reduced the kinetic energy of the flow. Hydraulic analysis indicated that flow intensity (Reynolds number) in zig-zag furrows was 1.4–1.6 times lower than in straight furrows (p< 0.05), leading to reduced turbulence and more stable flow conditions. These findings are consistent with the hydraulic mechanisms described by Trout et al. (1993) and Sepaskhah and Kamgar-Haghighi (1997).

Analysis of variance (ANOVA) showed statistically significant differences in seasonal irrigation water use between irrigation methods and slope gradients (p< 0.05) (Table 1). Under CSFI, total irrigation water input increased significantly with slope steepness, reflecting higher runoff losses and reduced infiltration efficiency. Seasonal water application ranged from 4,850 m³ ha^-1 at a 1.5° slope to 5,420 m³ ha^-1 at a 3.5° slope.

The application of ZZFI resulted in a consistent and significant reduction in seasonal irrigation water requirements across all tested slopes. Compared with CSFI, water use under ZZFI decreased by 640 m³ ha^-1 (13.2%) at a 1.5° slope, 695 m³ ha^-1 (13.6%) at a 2.5° slope, and 740 m³ ha^-1 (13.7%) at a 3.5° slope. On average, ZZFI reduced irrigation water use by approximately 15–20% relative to the control treatment, demonstrating high water-saving potential in sloping landscapes.

Nutrient dynamics were significantly influenced by the interaction between slope gradient and irrigation method (p< 0.05) (Table 2). Across all slope gradients, CSFI plots exhibited a substantial decline in soil organic matter and nutrient contents by the end of the growing season, primarily due to erosion-induced nutrient leaching. For instance, at a slope of 1.5°, total nitrogen content in the 0–30 cm soil layer decreased significantly from an initial value of 0.070% to 0.054% under CSFI.

In contrast, soils managed under ZZFI showed significantly smaller reductions in nutrient content (p< 0.05). At the same slope gradient (1.5°), total nitrogen declined only to 0.066%, representing a significantly better nutrient retention compared with the control. Similar patterns were observed for soil organic matter (humus), total phosphorus, and available phosphorus across the 2.5° and 3.5° slopes (Table 2). Crucially, at the steepest slope (3.5°), where erosion risk was highest, ZZFI plots maintained significantly higher end-of-season values for humus, nitrogen, and phosphorus fractions, confirming the method’s effectiveness in preserving soil fertility.

Soil erosion intensity showed a strong positive correlation with slope gradient under conventional irrigation practices (p< 0.05) (Table 3). Under CSFI, measured soil loss values increased sharply from 4.2 t ha^-1 at a 1.5° slope to 12.6 t ha^-1 at a 3.5° slope (Figure 2). This non-linear increase confirms that on steeper slopes (>2.5°), the kinetic energy of the flow in straight furrows exceeds the soil’s resistance to detachment, leading to severe degradation.

The implementation of ZZFI resulted in a drastic and statistically significant reduction in soil loss across all slope gradients (p< 0.01). Soil erosion under ZZFI was limited to 0.8 t ha^-1 at a 1.5° slope, 1.8 t ha^-1 at a 2.5° slope, and 3.5 t ha^-1 at a 3.5° slope. Consequently, ZZFI reduced total soil loss by approximately 4–5 times compared to the control treatment. This substantial mitigation is attributed to the reduced flow velocity and increased sediment deposition within the furrow channel, as observed in the hydraulic analysis.

Seed cotton yield was consistently and significantly higher under ZZFI compared to CSFI across all slope gradients (Figure 3). Under conventional irrigation (CSFI), crop productivity exhibited a strong negative correlation with slope steepness, decreasing from 3.45 t ha^-1 at a 1.5° slope to 2.85 t ha^-1 at a 3.5° slope due to soil degradation. In contrast, ZZFI plots maintained significantly higher yield levels (p< 0.05).

On average, ZZFI increased seed cotton yield by 0.32–0.43 t ha^-1 compared with CSFI. Water Use Efficiency (WUE) followed a similar trend. Under CSFI, WUE deteriorated with increasing slope, reaching 1902 m³ t^-1 at a 3.5° slope. The application of ZZFI significantly improved WUE across all slopes (p< 0.05), highlighting the method’s capability to produce more crop per drop of water (Table 3).

The results of this study clearly demonstrate that irrigation-induced soil erosion on sloping lands is primarily governed by hydraulic flow velocity rather than irrigation volume alone. Under conventional straight furrow irrigation (CSFI), increasing slope steepness resulted in a rapid acceleration of flow velocity, exceeding the critical threshold for soil particle detachment and transport. This finding is consistent with classical hydraulic studies, which indicate that once flow velocity surpasses a critical limit, shear stress increases disproportionately, leading to severe erosion (Trout et al., 1993; Trout, 1996).

By contrast, the zig-zag furrow irrigation (ZZFI) system fundamentally altered flow hydraulics by increasing the effective flow path length and surface roughness. This modification reduced flow velocity and turbulence intensity, as reflected by the lower Reynolds numbers observed in the zig-zag furrows. These results align with recent findings by Yu et al. (2022), who demonstrated that increasing channel obstruction and roughness significantly alters flow characteristics and reduces velocity. Similarly, Fernández-Gómez et al. (2004) emphasized that controlling inflow rates and channel geometry is critical for minimizing soil detachment on sloping terrains.

Importantly, our findings extend previous work (Sepaskhah and Kamgar-Haghighi, 1997) by demonstrating that hydraulic regulation through furrow geometry remains effective even at slope gradients as steep as 3.5°, where conventional systems typically fail. This suggests that ZZFI offers a robust erosion-control mechanism suitable for erosion-prone irrigated landscapes.

The substantial reduction in soil loss observed under ZZFI has direct implications for soil fertility preservation. It is well established that irrigation erosion selectively removes the finest and most fertile soil fractions, which are enriched in organic matter, nitrogen, and phosphorus (Keesstra et al., 2016; Pimentel and Burgess, 2013). In the present study, conventional irrigation resulted in pronounced declines in humus and total nitrogen contents, particularly on steeper slopes.

In contrast, zig-zag irrigation markedly reduced nutrient depletion, with nitrogen and phosphorus losses being 3–4 times lower than those recorded under CSFI. These results align with earlier regional studies reporting severe nutrient export from sloping irrigated fields under improper water management (Nurmatov and Umirov, 2003; Mirzayev, 2018). However, our study provides quantitative evidence that modifying furrow geometry alone—without altering fertilizer inputs—can significantly improve nutrient retention.

This finding is particularly relevant for meadow-sierozem soils, which are inherently low in organic matter and highly vulnerable to degradation. The ability of ZZFI to preserve soil chemical integrity underscores its role as an effective soil conservation strategy in arid irrigated systems.

Despite the demonstrated benefits, the widespread adoption of Zig-Zag Furrow Irrigation faces certain challenges. The primary limitation is the labor intensity required to form zigzag patterns manually in the absence of specialized machinery, as noted in the methodology. Furthermore, the mechanization of inter-row cultivation (weeding and fertilizing) is more complex in zigzag furrows compared to straight furrows.

Future research should focus on developing tractor-mounted implements capable of forming zigzag geometries automatically to reduce labor costs and enhance feasibility for large-scale farming. Additionally, long-term studies are needed to evaluate the cumulative effects of ZZFI on soil structural stability and salinization risks over multiple cropping cycles.

Water scarcity is intensifying under global climate change, particularly in arid and semi-arid regions where irrigation dominates agricultural water use (FAO, 2021). In this context, improving water use efficiency (WUE) without compromising crop yield is a key challenge. The present study demonstrates that ZZFI substantially reduces seasonal irrigation water requirements while simultaneously increasing cotton yield.

The observed 15–20% reduction in water use under ZZFI can be attributed to lower runoff losses and enhanced infiltration resulting from reduced flow velocity and increased wetted perimeter. Similar water-saving effects have been reported for alternative furrow irrigation strategies, such as surge flow, in the Fergana Valley (Horst et al., 2007). However, many of these approaches require additional infrastructure or operational complexity.

In contrast, ZZFI achieves water savings through a simple structural modification of existing furrows. The simultaneous increase in yield and reduction in water input resulted in a significant improvement in water use efficiency, illustrating the practical applicability of the “more crop per drop” principle in sloping irrigated fields.

From an ecosystem services perspective, the results indicate that zig-zag furrow irrigation enhances regulating ecosystem services while supporting provisioning services (Figure 4). Reduced soil erosion and nutrient losses directly strengthen soil conservation, nutrient cycling, and water regulation functions, which are essential for long-term agroecosystem resilience (Lal, 2015; Keesstra et al., 2016).

At the same time, higher and more stable cotton yields under ZZFI demonstrate that improved regulating services do not come at the expense of agricultural productivity. Instead, the zig-zag approach reconciles the often-conflicting goals of environmental protection and crop production. This balance is particularly critical for Central Asian irrigated systems, where soil degradation and water scarcity pose simultaneous challenges (Isaev et al., 2020).

By integrating erosion control, nutrient retention, and water efficiency into a single low-cost practice, ZZFI aligns closely with the principles of conservation agriculture and sustainable land management.

The practical relevance of zig-zag furrow irrigation lies in its simplicity and adaptability. Unlike drip or sprinkler systems, which require substantial capital investment and energy inputs (Chartzoulakis and Bertaki, 2015), ZZFI can be implemented using existing irrigation infrastructure with minimal additional cost.

Because large areas of irrigated land in Uzbekistan and other parts of Central Asia are on sloping terrain, adopting ZZFI could significantly reduce soil degradation and improve water productivity on a regional scale. This technology is ideal for small and medium-sized farms where financial and technical limitations restrict access to sophisticated irrigation systems.