We found that the window phase for the total N loss mainly appeared in the growing season of rice. However, the window phase for the total P loss was more dispersive. This may be due to the different behavior of N and P in the runoff and the management of the fertilizer for the planting of rice. The demand for N in rice is much higher than that for P. In this experiment, N fertilizer (186 kg N ha− 1) was applied to the soil on three occasions, mainly in July-August. However, P fertilizer (120 kg P2O5 ha− 1) was applied as a base fertilizer before rice planting. In addition, studies have shown that the soil P has a much lower migration rate than the soil N . Therefore, the window phase of N is mainly concentrated in the rice-planting season. The window phase of P is relatively dispersed throughout the year, which is related to the surface runoff. The study showed that the peak flows in the window phase usually contained the majority of N and P loss [38, 39].
Meanwhile, the loss window phases of N and P are consistent with the runoff pattern. When runoff occurs, greater P loss loading also occurs (Fig. 4). However, the concentration does not show the same pattern, and the concentration of P in the runoff during heavy rainfall periods (especially the rice-planting season from June to July) is rather low. This may be due to the strong effect of P immobilization and the adsorption onto the soil. Some studies have shown that the P in the paddy soil is mainly organic P and inorganic Al oxide-P and Fe oxide-P , which is wrapped in aggregates and is not easily released into runoff. Some studies showed that phosphorus migrates mainly through small-sized organic-inorganic composite colloids [41, 42]. Therefore, its mobility is relatively weakly affected by runoff, and its concentration increases when soluble P and colloidal P enter the water body with increasing runoff amount. When the rainfall is particularly heavy, all the soluble P and colloidal P in the soil are transferred into runoff, and the concentration is diluted, resulting in the lower concentration. However, owing to the fertilization season during the rice-planting period, the N concentration in the soil is remarkably high, and the peak value of the N concentration is consistent with the peak value of the heavy rain during this period. Lang found a significant and positive correlation between the soil depth and the flow-weighted total N and P concentrations . Moreover, the results of their study revealed that the irrigation water N and P concentrations have an impact on the drainage N and P loads. However, the total N and P transport to the tile-drains was primarily regulated by the macropore flow and was not significantly affected by storms . Furthermore, a study on the nutrient losses from a heavy clay soil in a fluvial plain in the Netherlands found that the contribution of pipe drains to the total discharge of N and P was strongly related to the length of the dry period in the preceding summer because of the very low conductivity of the soil matrix and the formation of shrinkage cracks . Another study reported that the P concentration in the drainage flow was independent of the Olsen P, while the antecedent soil moisture had a strong influence on the P loss in the drainage flow. Furthermore, higher concentrations were recorded above a certain soil moisture deficit threshold .
The Δtotal N (drainage minus stream) showed negative values, while the Δtotal P showed positive values, which also indicated the different role of steam on N and P at different periods. The stream water quality is often influenced by field runoff or channel drainage. A study on runoff in the Tipton Creek watershed (Iowa) showed that the channel drainage dominated and was an important source of the total N and P discharge of the watershed . In the present study, high total P concentrations were observed in the water samples, especially in the water of the retention ditch. More than 75% of the drainage water samples had greater P concentrations than that stated in the United States Environmental Protection Agency (US EPA) guidelines (0.10 mg total P L− 1 for streams) , and more than 90% of the samples exceeded 0.02 mg total P L− 1, which is a critical total P value, above which the eutrophication in the surface water would accelerate . These data indicate that the water monitoring and forecasting of the retention ditch are extremely necessary to protect the water quality of the adjacent streams.
We also found that the N and P loss loadings, rather than the concentrations of N and P in the retention ditches, were linearly related to the surface runoff, which confirmed our hypothesis. This result shows that we can predict the loss loadings of N and P by the cumulative runoff amount. As the N and P concentrations fluctuate across the rice seasons, the N and P loss loadings cannot simply be calculated by multiplying the concentration by the runoff at each time point. This relationship provides a simple tool for predicting the nutrient loss loading in the retention ditches of the paddy field. A study in the Baltic Region also showed that no correlations between the nutrient concentrations and precipitation were found in the Mrzezino canal . Meanwhile, other studies have indicated the similarities in P loss in the tile drainage and surface runoff, as well as the strong correlations between P loss and storm hydrographs .
Using high-resolution monitoring at 6 h intervals, we found that the rainfall amount was an important factor that affects the total N and P concentrations in the retention ditch. These findings are supported by the results of other studies that were conducted in eastern China and other Asian monsoon areas [50, 51]. However, the SOM did not reveal a better relationship between the rainfall at 6, 12, 18, 24, 36, 48 h and the N and P concentrations. Some studies involving high temporal resolution monitoring (at 0.5 h intervals) have shown that using a 1 h sampling frequency might lead to a higher error in assessing the total pollution load than that obtained with a 0.5 h frequency . However, determining if this error is an over- or under-estimation depended on which 1 h sample set was used. This was also consistent with our result, which revealed that using more than 24 h monitoring data did not produce improved results over using shorter monitoring periods with regard to the relationship between the total N and P in the retention ditch and the accumulated rainfall amount.
Nevertheless, we believe that our SOM approach can perform better in future studies if additional environmental and management factors are considered. In addition to the rainfall, there are many other factors, such as soil conditions (soil moisture, soil depth, soil macropore, available N and Olsen P concentrations) and management practices (irrigation N and P concentrations, dry period, land cover, and fertilization), that influence the drainage water N and P concentrations or loadings [4, 43, 53]. The SOM approach developed in this study helps to predict the total N and P in the drainage based on high-resolution monitoring at 6 h intervals. These approaches were easy to implement by river managers in designing a long-term protection plan to maintain the water quality; however, it may be unable to predict the daily total N and P concentrations based on the daily rainfall forecast. In the present study, we only investigated the total P and N in the drainage water. Other P and N forms, especially particulate, colloid, soluble, reactive, and unreactive P, NH3+-N, and NO3−-N should be further tested, because several studies have shown that their response to rainfall in storm events was much closer than that of the total P .