Many factors influence the mobilization and transport of non-point source pollutants from soils to water bodies including rainfall regimes, land use and land-management practices, and topography. Each of these factors is considered in the following section, with a focus on their implications for soil erosion-type NPSP.
3.1 Rainfall regimes
Precipitation is a prerequisite for runoff, and rainfall intensity and duration primarily control runoff generation and, consequently, associated erosion and nutrient mobilization (Zhang et al., 2010). For example, in the Erie Lake Basin, North America, Williams et al. (2020) showed that rainfall intensity and duration are important factors affecting phosphorus loss dynamics. In addition, Wang et al. (2014) and Ding et al. (2017) also demonstrated that rainfall intensity is an important factor affecting phosphorus loss.
During an individual rainfall event, N&P loss typically increases first and then gradually decreases and stabilizes over time (Wang 2019b). Importantly, runoff generation times are closely related to rainfall intensity; at higher rainfall intensities, runoff generation time advances with associated increases in runoff rates and N&P losses. For example, using rainfall simulation experiments, Wang (2019b) showed that rainfall intensities of 50, 75, and 100 mm/h produce runoff generation times of 1.5 min, 1.2 min, and 0.5 min, respectively. Furthermore, the same authors show that with increasing rainfall intensity, the loss of TN in surface runoff increased from 214.55 to 1,017.02 mg and the loss of TP increased from 5.63 to 48.85 mg.
Rainfall (such as rainfall intensity and rainfall duration) is a direct factor leading to the migration and loss of nutrients from soil, which is significantly positively correlated with TN, TP, soluble nitrogen, and soluble phosphorus. For example, Li et al. (2013) observed that pollutant (TN) losses via surface runoff from chestnut forests in the Fushi Reservoir catchment in Anji County, Zhejiang Province, China, were significantly positively correlated with rainfall.
3.2 Impacts of land use and production methods on N&P loss
Farming methods affect the loss of N&P from soils by changing runoff generation times and intensities. For example, on the Sanjiang Plain in China, Huang et al. (2015) showed that organic phosphorus loss from drylands, rice fields, forested land, and wetlands was 2.30, 2.04, 0.99, and 0.69 kg/ha, respectively, and TP loss was 3.28, 3.04, 1.43, and 1.04 kg/ha, respectively. Yang et al. (2018) studied the loss of nitrogen and phosphorus under different farming modes (CNL-CK, corn + no straw application + longitudinal ridge; CSSC, corn + stubble standing + cross ridge; CSC, corn + straw application + cross ridge; SSC, soybean + straw application + cross ridge; ASC, alfalfa + straw application + cross ridge) in the Songhua River area. Compared with the traditional CNL, CSSC could reduce 53.8% soil nitrogen loss, 50.7% soil phosphorus loss, 52.8% runoff nitrogen loss and 46.9% runoff phosphorus loss. CSC could reduce 50.8% soil nitrogen loss, 50.7% soil phosphorus loss, 56.8% runoff nitrogen loss and 78.2% runoff phosphorus loss. SSC could reduce 62.1% soil nitrogen loss, 60.0% soil phosphorus loss, 46.1% runoff nitrogen loss and 67.2% runoff phosphorus loss. ASC could reduce 66.2% soil nitrogen loss, 66.7% soil phosphorus loss, 50.2% runoff nitrogen loss and 62.7% runoff phosphorus loss.
The type of crops planted in farmland soils has a direct influence on N&P losses. For example, in the Yujiahe River Catchment in Shaanxi, China, Chen et al. (2019) showed that while farmland only accounts for 28% of the land area, this accounted for 80.4% of the nitrogen loss in the catchment, and orchard soils accounted for 66.7% of the loss of available phosphorus. Furthermore, Wang et al. (2019d) conclude through simulation experiments in Henan Province, China that farmland is the land use most conducive to N&P loss followed by tea gardens and citrus orchards, while losses from forested land are comparatively low.
Most fertilizers contain N&P and, therefore, their application directly alters the concentrations of non-point source pollutants in agricultural soils. Bouraima et al. (2016) studied the loss of nitrogen and phosphorus under four fertilization methods (CK, without fertilizer; T1, combined manure with chemical fertilizer; T2, chemical fertilization; T3, chemical fertilizer with increasing fertilization) in Chongqing for five years (2010-2014). The study showed that compared with CK, T1 treatment can reduce 41.2% of TN loss and 33.3% of TP loss. In their study of 62 watersheds in Minnesota, USA, Boardman et al. (2019) found that fertilizer application at the crop-planting stage was the main cause of nitrogen pollution. Zhang et al. (2019) studied in Jialing River showed that N&P loss is positively correlated with fertilizer application rates; for every 10% increase in fertilizer application, N&P loss increased by 1% and 4%, respectively. Across China, Zou et al. (2020) showed that the application of chemical fertilizers has increased from 8.84 million tons in 1978 to 58.59 million tons in 2017, and the application of pesticides has also increased from 0.73 million tons in 1990 to 1.66 million tons in 2017. It can be seen that choosing the appropriate type and amount of fertilizer and applying it at the appropriate time are crucial to reducing water pollution.
3.3 Terrain
Topography determines the redistribution and intensity of runoff, which have a significant impact on slope erosion and nutrient transport (Zhao et al. 2016). For example, Wu et al. (2018) conducted artificial simulated rainfalls and found that as the slope rises, the dissolved TP lost with surface runoff increases significantly. Furthermore, in Tianmu Lake, East China, Zhang et al. (2020) report that variations in TN (7.0–10.0 mg/L) and TP (0.05–0.07 mg/L) concentrations entering the lake correspond to variations in the gradient of the surrounding slopes (8–16°). These studies also showed that topography (slope) is closely related to the loss of non-point source pollutants such as nitrogen and phosphorus. Bai et al. (2020) studied the characteristics of nitrogen and phosphorus loss in a special soil (degraded Ferralsols) under different slopes (10°, 15°, 20°) in Fujian, China. They found that when the slope is 15°, it is most beneficial to reduce the loss of nitrogen and phosphorus.