Best management practices for reducing nutrient loads under diverge hydrological conditions: A case study in an agricultural watershed in Southwest China

• We assessed nutrient reduction rates achieved by Best Management Practices (BMPs). • Scenario analysis shows up to 32% nitrogen and 43% phosphorus can be reduced. • Fertilizer reduction can effectively reduce nitrogen during flood season; • Water storage ponds can effectively reduce nitrogen during the non-flood season. • Phosphorus can be effectively reduced mainly due to no-tillage and filter-stripe. Best management practices (BMPs), such as no-tillage and filter strips, have the potential to reduce agricultural non-point source pollution and enhance water quality. However, their effectiveness was not well understood, particularly under diverging hydrological conditions. Our study evaluated the effects of BMPs on nutrient load reduction at the watershed outlet under varying hydrological conditions and water quality targets, using the Fengyu River watershed as a case study. We applied the SWAT (Soil and Water Assessment Tool) model to quantify nutrient loads at the watershed outlet under different BMP scenarios from 2011 to 2013. Additionally, we used the load duration curves method to evaluate the nutrient reduction effectiveness of these BMP scenarios compared to the baseline scenario under different hydrological conditions. Results showed that BMP scenarios reduced the annual total nitrogen (TN) load by 1–32 % and total phosphorus (TP) load by 4–43 % at the watershed outlet relative to the baseline in 2011–2013. Among them, FR-II (20 % fertilizer reduction) and PDS (ponds) were more effective in reducing annual TN load at the watershed outlet by 14 % and 15 %, respectively, while NTL (no-tillage) and FS-III (5-meter filter strips) were more effective in reducing TP load by 12 % and 30 %, respectively. The load duration curve analysis demonstrated that FR-II and PDS were effective in reducing TN load under both high and low streamflow conditions compared with other individual BMPs. NTL and FS-III were more effective in reducing TP under high streamflow conditions. To maximize nutrient reduction effectiveness, a combination of BMPs is necessary. BMP combinations tailored to different streamflow intervals significantly increased the percentage of daily nutrient flows that achieve water quality targets at the watershed outlet. Our study provides new insights into the effectiveness of BMPs and supports watershed managers in formulating strategies to reduce nutrient pollution and improve water quality.