Experimental Site

The study was carried out at a typical irrigation area of the Hangjiahu Plain River Network in Zhejiang Province (30°35′N, 120°57′E) from May to October in 2020 and 2021. The study area belongs to a monsoon climate zone. The average annual temperature is 15.7℃, the average annual sunshine hours are 2075 hours, the annual rainfall is 1252.4 mm, the average frost-free period is 224 days, and the soil is mainly composed of silty clay with a unit weight of 1.3–1.4 g/cm³. The irrigation area is 36.67 ha, with rice (conventional single-cropping rice Xiushui 12) as the planting crop. There is one irrigation pump station, a low-pressure irrigation pipeline of 5.11 km, a drainage soil ditch and U-shaped concrete ditch of 2.77 km, and an ecological drainage ditch of 0.52 km. A natural wetland system has been constructed by utilizing existing rivers and ponds near the main drainage outlet.

Experimental Design

This study was based on the scale of irrigation areas, taking farmland, drainage ditches, and ponds (wetlands) as a whole. Coupling water-saving irrigation and controlled drainage technology was adopted in farmland irrigation and drainage regulation, the core of which was to reduce the upper limit of irrigation and increase the upper limit of rainwater storage. The control standards for farmland irrigation and drainage are shown in Table 1. The N and P pollutants came from fertilization in rice fields, which was applied with base fertilizer, tillering fertilizer, and panicle fertilizer, and the fertilization time was before the tillering stage, during the tillering stage, and before the heading and flowering stage during the rice growth period, respectively. Fertilizer varieties and amounts are shown in Table 2.

The drainage ditch was divided into three levels, they were farm ditches (soil ditches), branch ditches (U-shaped concrete ditches, ecological ditches), and main ditches (river channels). The farm ditches were distributed along the field with a spacing of 10 m, a width of 0.5 m, a depth of 0.8–1.0 m, and a daily drainage depth of 0.4 m. The width of the branch ditch was 1.0 m, the water depth was maintained at 1.0–1.5 m, and there was a regulating gate at the intersection with the river channel. The main ditches were regional river channels. In this study, three kinds of drainage systems were set up, namely ditch-sluice gate regulation and storage system (DSG system), field-ditch direct drainage system (FDD system), and ecological ditch-pond regulation and storage system (EDP system). On the basis of direct drainage in the irrigation area, the DSG system set up a gate at the outlet of the farm ditch to control the water level in the irrigation area; when the water level reached the upper limit in farm ditches, it would drain in 3–4 days later. FDD system adopted open channel drainage measures, which discharge excess surface water through a drainage system composed of excavated open channel channels, including a multi-level channel drainage system. The EDP system used both ecological ditches drainage and ditch-ponds (wetlands) for regulation and storage, which were arranged with aquatic plants. The upper limit of water storage was 20–50 cm below the lowest field elevation in the drainage range, the water storage reached the lowest field elevation in the drainage range after a rainstorm, and the suitable drainage period was 7–8 days. The layout of the drainage system in the study area is shown in Fig.1.

Indicators and Measurements

Water monitoring included irrigation water consumption monitoring and drainage monitoring. The irrigation water consumption monitoring was carried out in real-time by using flow meters and field water meters located at the head of the pump station and the inlet of typical fields. The drainage amount was monitored in real-time by a simple water measuring device located at the outlet of the end of the drainage ditch. Water quality sampling points were set up at the pump station inlet, fields, farm ditches (soil ditches), U-shaped concrete ditches, ecological ditches, and ponds (wetlands) to monitor the water quality changes. Water samples were taken every 7 days and on the 1^st, 2^nd, 3^rd, 5^th, and 7^th days after heavy rainfall or fertilization. The monitoring indicators included Total nitrogen (TN), Total phosphorus (TP), NH₄⁺-N, NO₃^--N. TN was measured using potassium persulfate oxidation and ultraviolet spectrophotometry. TP was measured in unfiltered samples using the indophenol blue method. NH₄⁺-N was determined by the Nessler reagent spectrophotometry method, and NO₃^--N concentration was measured by the Ultraviolet spectrophotometry method. State of Environmental Protection Association (SEPA, 2002) was adopted for measurement methods of TN, TP, NO₃^--N, and NH₄⁺-N.

Statistical Analysis

Removal Rate Calculation

The removal rate was calculated according to the following formula.

η=∑qinflow−∑qoutflow∑qinflow

qinflow/outflow=Qc

Where η represented the removal rate (%), q_{inflow/outflow} represented the amount of material inflow and outflow (g/ha), Q represented the displacement (m³/ha), and C represented the concentration of pollutants (mg/L). A positive value indicated that the total inflow of substances was greater than the total outflow of substances, resulting in an increase in the total amount of pollutants. A negative value indicates that the total inflow of substances was less than the total outflow of substances, resulting in a decrease in the total amount of pollutants, and zero indicates no inflow (outflow).

Fig. 1.

Schematic diagram of drainage system layout in the study area.

Temporal and Spatial Changes of N and P Pollutants in Different Drainage Systems

Field-Ditch Direct Drainage System

Changes in N and P pollutants concentrations in various levels of drainage ditches of FDD system are shown in Fig. 3. The concentration ranges of TN, NH₄⁺-N, NO₃^--N and TP were 0.63–5.54 mg/L, 0.06–1.59 mg/L, 0.068–2.37 mg/L and 0.11–0.514 mg/L, and N and P pollutants concentrations were gradually decreased along farm ditches, branch ditches, and main ditches. The NH₄⁺-N changes showed an overall downward trend in 2020 and 2021. The changes in NH₄⁺-N concentration in various levels of ditches were similar in 2020, but in farm ditches, they were relatively large in 2021.

The NH₄⁺-N concentration in each ditch generally increased in July and August due to fertilization and topdressing but showed a trend of first increasing and then decreasing in September, mainly due to the vigorous growth of crops, large nitrogen absorption and reduced pollutant load output from farmland. For the NO₃^--N changes, it showed an overall downward trend in 2020 and 2021, but it changed significantly in 2021.

The NO₃^--N concentration was relatively high from June to August in 2020 and 2021, which may be due to the favorable conditions for nitrification reaction under aerobic conditions, resulting in an increase in NO₃^--N concentration. For the TN changes, the decrease in TN concentration in 2021 was higher than that in 2020, and it fluctuated greatly in 2021. Different from the DSG system, the average concentrations of TN in branch ditches and farm ditches were similar in the FDD system. For the TP changes, it was the smallest pollutant compared to N pollutants, with little change in 2020 and a fluctuating downward trend in 2021. After heavy rainfall occurred in September 2020 and 2021, the TP concentration significantly increased because rainfall was the carrier of soil solutes and the driving force for phosphorus migration in paddy soil.

Fig. 2.

Changes in N and P pollutants concentrations in various levels of drainage ditches of DSG system.

Fig. 3.

Changes in N and P pollutants concentrations in various levels of drainage ditches of FDD system.

Absorption and Removal Capacity of N and P Pollutants in Different Drainage Systems

Removal Effect of N and P Pollutants

The inflow, outflow, and removal rate of pollutants under different drainage systems are shown in Table 3. It showed that the DSG system had the highest removal rate of TP, which was 59.5% and 8.0% higher than the FDD system and EDP system. The removal rate of TN in the EDP system was the highest, with an increase of 35% compared to the DSG system and FDD system. The EDP system had the highest removal rate of NO₃^--N, which was significantly improved compared to the other systems, while the DSG system had the lowest removal rate. The removal rate of NH₄⁺-N in the EDP system was the highest, followed by the DSG system. Overall, each drainage system had a certain retention and purification effect on N and P, and its purification capacity from priority to inferiority was in the order of the EDP system, DSG system, and FDD system. The better removal efficiency of the DSG system was mainly due to the equipped control gate in the farm ditch, which was longer than that of the FDD system. The farmland drainage was retained in the farm ditch for a longer time, and the flow rate was reduced, which was conducive to the physical settlement and filtration of pollutants. In the EDP system, the application of fertilizer in the field resulted in a higher content of nutrients such as N and P that had not been fully utilized, leading to a higher concentration of N and P pollutants flowing into the system. However, the farmland drainage passed through the ditches and ponds, and there was a longer drainage ditches system that provided plants and microorganisms with sufficient interception and digestion of the water body, resulting in better purification of the water body. Additionally, due to the vigorous growth of aquatic plants, developed root systems, and high temperatures within the EDP system, microbial metabolic activities were strengthened, which was conducive to the full absorption and utilization of aquatic plants and microorganisms and the reduction and purification effect of N and P pollutants was significant, resulting in a higher pollutant removal rate and better retention and purification effect. Meanwhile, compared to N, the P removal rate was higher, which may be influenced by aquatic plants and substrate conditions.

Fig. 4.

Changes in N and P pollutants concentrations in various levels of drainage ditches of EDP system.

Table 3.

Inflow, outflow, and removal rate of pollutants under different drainage systems.

Drainage system	Indicators	TP	TN	NH4+-N	NO3--N
DSG system	Pollutants inflow (g/ha)	334.5	2298	950	942
	Pollutants outflow (g/ha)	253.5	1872.5	902.5	924
	Difference (g/ha)	81	425.5	47.5	18
	Removal rate (%)	24.2	18.5	5.0	1.9
FDD system	Pollutants inflow (g/ha)	253.5	1872.5	902.5	924
	Pollutants outflow (g/ha)	252	1606.5	862	771
	Difference (g/ha)	1.5	266	40.5	153
	Removal rate (%)	0.4	14.2	4.5	16.2
EDP system	Pollutants inflow (g/ha)	409.5	3138	1128	1827
	Pollutants outflow (g/ha)	334.5	2298	950	942
	Difference (g/ha)	75	832.5	178	885
	Removal rate (%)	22.4	36.2	15.8	93.9

Fig. 5.

Changes in N and P pollutants concentrations at the inlet and outlet in different drainage systems.

Fig. 6.

Variations of pollutant removal rate with degradation coefficient in ditches and ponds under different hydraulic conditions.

Effect of Hydraulic Conditions on the Removal of N and P Pollutants in Ditch-Pond Systems

Variations of pollutant removal rates with degradation coefficient in ditches and ponds under different hydraulic conditions are shown in Fig. 6. The ideal removal rate and actual removal rate both increased with the increase of pollutant degradation coefficient, and the pollutant removal rate of the ditch-pond system under actual agricultural hydraulic conditions was only equivalent to the overall consideration of 65.6~91.1%. The implementation rate showed a flat U-shaped trend with the increase of pollutant degradation coefficient when the degradation coefficient was very small (0.01 d^-1), ε was about 0.75 when the degradation coefficient increased to 0.07 d^-1, ε decreased to the minimum value of 0.66, and then slowly increased with the increase of degradation coefficient, when the degradation coefficient increased to 0.30 d^-1, ε slowly increased to 0.8. In addition, when the degradation coefficient was small, improving the hydraulic connection of the ditch-pond system had little effect on pollutant removal. However, as the degradation coefficient increased, the impact of hydraulic connection on pollutant removal efficiency gradually increased. Therefore, improving the purification effect of drainage water quality can be approached from the perspective of increasing the degradation coefficient of pollutants. Correspondingly, under the calculation conditions, the actual removal rate of pollutants was also greatly affected. When the degradation coefficient was very small (0.01 d^-1), the pollutant removal rate obtained by considering hydraulic connections was 5%, while the ideal situation was 6% when the degradation coefficient increased to 0.1 d^-1, the pollutant removal rate obtained by considering hydraulic connections was 32%, and the ideal maximum value was 46%, and when the degradation coefficient increased to 0.3 d^-1, the actual removal rate of pollution was only 78%, while the ideal removal rate was 86%. Within the range of degradation coefficients proposed in existing research, if calculated according to ideal conditions, pollutants would be significantly reduced, while after considering specific hydraulic connections, the pollutant removal rate would decrease. Therefore, the pollutant removal rate of ditches and ponds in their natural state was significantly lower than that after optimization, indicating that improving hydraulic connections were crucial for enhancing the pollutant degradation capacity of ditches and ponds.

Factors Affecting the Removal of N and P Pollutants in Different Drainage Systems

In the agricultural drainage ditch system, N and P nutrients are mainly intercepted and purified through sediment adsorption, absorption by aquatic plants, and microbial metabolism, reducing the amount of pollutants entering rivers and lakes [21]. This paper studied the spatiotemporal changes of N and P pollutants in different drainage systems, including main ditches, branch ditches, and farm ditches. It was found that the pollutant concentration in various levels of ditches would fluctuate significantly under different periods and meteorological conditions. On the whole, the concentration of N and P pollutants showed a consistent change, with higher concentrations in July, greater fluctuations in August, and the lowest in September and October The result was consistent with the research of Gao [22]. However, Fang found that the concentrations of NH₄⁺-N, TN, and TP in fall and winter were significantly higher than those in other periods, which might be related to differences in meteorological conditions in the study area [23].

There are significant differences in the variation of non-point source pollutants at different levels of drainage systems, as agricultural drainage ditches at different scales had different structural dimensions, internal environmental characteristics, and differences in the concentration of incoming pollutants [24]. For the DGS system, the peak concentration of various pollutants occurred in the farm ditches, while the concentrations of pollutants in branch ditches and main ditches were relatively low. It was because there was a regulating gate at the end of the farm ditch. Under the combined effect of fertilizer migration and rainfall, the water flow rate was slow, and the discharge water from farmland stayed in the ditch for a longer time, which was conducive to the physical settlement of pollutants and the full absorption and digestion of aquatic plants. For the FDD system, the average content of TP gradually increased, that was because the phosphorus fertilizer was easy to fix after entering the soil, and it was generally fixed in the topsoil layer, so it was difficult to enter the deep soil during the irrigation season, the soil turbulence was small, and the TP concentration was changed little, but after the rainstorm, the soil was hit, so the TP in the topsoil entered into the water and moved to drainage ditches, and at the same time, the water flow in the ditches was increased, and the continuous erosion of the rain made the phosphorus element difficult to transform. For the EDP system, NH₄⁺-N and TN concentrations were higher in farm ditches but significantly lower in the main ditches; that was because the EDP system played an important role in interception and purification under the absorption of plants, microorganisms, and other effects, and it had strong resistance to impact loads.

This study mainly considered the effects of rainfall and fertilization on the spatiotemporal changes of surface source pollutants at various levels of the system. However, Wang found that factors such as hydraulic retention time, plant species, biomass, temperature, hydraulic load, and wet-dry alternation were also influencing factors in reducing non-point source pollutants in farmland [25]. Currently, there is relatively little research on the impact analysis of relevant factors on the reduction effect of N and P pollutants in farmland. Attention still needs to be paid to the mechanism of hydrological and environmental factors on the removal effect of N and P pollutants in drainage ditches in order to obtain more theoretical support.

Impact of Drainage Systems on the Digestion ability of N and P Pollutants

This study found that during the monitoring period, the removal rates of N and P pollutants by the drainage systems were all positive. Among them, the EDP system had the best effect, while the FDD system had a relatively poor impact. It was because agricultural non-point source pollution mainly came from the use of fertilizers and pesticides, and N and P pollutants that were not absorbed and utilized by crops would be directly discharged into the river channel, which would have a significant impact on the surrounding water quality [26]. The EDP system had the highest removal rate of NO₃^--N, followed by TN. Ecological ditches, as a special wetland system, had numerous aquatic plants such as Acorus calamus, reed, and water celery growing in the drainage ditches. The plants grew rapidly and absorbed a large amount of N and P pollutants to achieve the effect of dissolving and removing pollutants. At the same time, plant roots, sediment, and humus formed by the decay of aquatic plant epidermal cells had a strong adsorption effect on pollutants, especially organic matter, to achieve the effect of intercepting and purifying pollutants. The removal efficiency of TN and TP in the EDP system had been significantly improved through wetland ditches and ponds. Therefore, the combination of ecological ditches and wetland ponds had a significant effect on N and P removal [27]. Wang found that plankton in ecological ditch storage systems also had assimilation and absorption effects on N and P elements [28]. In addition, for agricultural drainage, multiple ecological defense lines, such as rice paddies, drainage ditches, and wetlands, can be used to absorb and reduce N and P nutrients in order to reduce the discharge of non-point source pollution. At the same time, the types of aquatic plants in the ecological ditch should be enriched, and aquatic plants should be chosen that are less prone to decay in winter to reduce the probability of secondary pollution caused by aquatic plants to the water body. In terms of the structural form of ecological ditches, it was possible to combine local conditions and adopt structural types that were suitable for the surrounding environment in order to maximize the reduction of non-point source pollutants.

Efficient Technology for Removing N and P Pollutants at the Scale of a Ditch-Pond (Wetland) System

Surface drainage is an important way for N and P from rice fields to enter the receiving water body. The water discharged from farmland through drainage ditches, ponds, and wetlands could further regulate and store some water and effectively reduce N and P nutrients. Therefore, the collaborative regulation field-ditch-pond is adopted to connect the rice field, drainage ditch, and pond as a whole [29]. Rainfall is initially intercepted by the farmland, and the drainage ditch and pond (wetland) system intercept the farmland drainage twice or multiple times, fully exerting the wetland effect of field ditch-pond to reduce N and P concentration, forming a regulation model of the paddy field-ditch-pond (wetland) system (Fig. 7). This regulation mode can be divided into three stages, among which stage one is field controlled drainage, that is, during the growth period of rice with strong resistance to waterlogging stress, the maximum water storage depth after rain can be controlled to 150–200 mm, rice fields do not actively drain water, when the waterlogging occurred in the rice field exceeds 5 days, it should be drained, and during fertilization and pest control, the waterlogging can be extended to 7 days before being discharged [30]. The second stage is farm ditch storage and drainage, where the upper limit of farm ditch water storage is the elevation of the gate top, which is the elevation of the field surface. After the water storage in the farm ditch reaches the upper limit, it will be emptied 3–4 days later, and when there is moderate to heavy rain or fertilization, the farm ditch will be emptied in advance [31, 32]. The last stage is pond (wetland) storage and drainage. The upper limit of pond (wetland) storage is 20–50 cm below the lowest field surface elevation in the range of bearing and discharging. When a rainstorm occurs, it would be stored to the lowest field surface elevation in the range of bearing and discharging, and when the upper limit is reached, it will be necessary to accommodate the drainage of water from upstream again, the suitable drainage period is 7–8 days, and it will be discharged to the same level as the downstream [33].

Fig. 7.

Schematic diagram of an ideal regulation mode for field-ditch-pond (wetland) system.

In summary, in actual production, on the one hand, it is necessary to improve the construction of drainage ditches at all levels in irrigation areas. On the other hand, irrigation areas should retain a certain amount of pond water surface (or wetland) and regularly carry out dredging to ensure the rational utilization of rainwater resources and efficient removal of non-point source pollution from farmland.