赤红壤蔗区11年连续增量施磷下磷素演变及其 对甘蔗产量与磷流失的影响

【目的】 系统分析连续11年增量施磷下赤红壤蔗地土壤全磷、Olsen-P以及地表径流磷流失量的变化特征和土壤磷素变化与磷盈亏、蔗茎产量的响应关系,为土壤磷素科学管理提供参考。【方法】 依托长期肥力及地表径流定位监测试验（2008年—）,选取不施肥（CK）、推荐施肥（OPT）和增量施磷（OPT+P）3个处理,测定土壤全磷、Olsen-P含量及地表径流磷流失量,分析土壤磷素变化与磷累积盈亏量的关系,采用Mitscherlich模型拟合蔗茎产量对Olsen-P的响应曲线,计算土壤Olsen-P农学阈值,并推算施肥处理土壤Olsen-P含量从第11年降至环境阈值所需的时间。【结果】 CK处理逐年降低土壤全磷含量,年降速率为0.0251 g·kg ^-1·a ^-1。施肥土壤全磷和Olsen-P含量随种植年限波动增加,土壤全磷和Olsen-P增速率OPT+P处理高于OPT处理。不施肥土壤表观磷盈亏10.2 kg·hm ^-2·a ^-1,施肥处理土壤表观磷盈余41.3—69.2 kg·hm ^-2·a ^-1,占施磷量的31.9%—35.6%,以OPT+P处理显著高于OPT处理67.5%。施肥下赤红壤蔗区土壤全磷和Olsen-P变化量均与土壤累积磷盈亏量呈显著正相关关系（P＜0.01）,土壤每累积盈余100 kg P·hm ^-2,OPT处理和OPT+P处理土壤全磷上升0.06 g·kg ^-1和0.09 g·kg ^-1,Olsen-P 含量上升11.0 mg·kg ^-1和9.1 mg·kg ^-1。土壤每累积亏缺100 kg P·hm ^-2,CK处理土壤全磷下降0.32 g·kg ^-1。Mitscherlich模型较好地拟合蔗茎产量与赤红壤Olsen-P含量的响应关系（P＜0.01）。其计算出的土壤Olsen-P 农学阈值为12.1 mg·kg ^-1。施肥显著提高地表径流磷流失量,且OPT+P处理也显著高于OPT处理。地表径流磷流失量与土壤Olsen-P含量显著正相关。基于土壤磷素变化与累积磷盈亏的关系推算得出第11年OPT和OPT+P处理Olsen-P水平降至环境阈值的时间分别需要12年和16年。 【结论】 在南方赤红壤区,施肥尤其增量施磷在提高土壤磷素累积的同时增加了地表径流磷流失风险。在本试验磷的基础养分条件下,按OPT处理施磷,并从甘蔗种植的第2—3年实行隔年施磷可维持土壤磷素处于农学阈值与环境阈值之间。

Change of Phosphorus in Lateritic Red Soil and Its Effect on Sugarcane Yield and Phosphorus Loss in Runoff Under 11-Year Continuous Application of Excessive Phosphorus Fertilizer

【Objective】This study was conducted to explore the change of total Phosphorus (P), Olsen-P and P loss in runoff in lateritic red soil, and to evaluate the relation between P change and P balance and between P change and cane yield under 11-year continuous excessive P fertilization, with an aim to provide scientific references for scientific management of soil P.【Method】Three different fertilization treatments (non-fertilization (CK), optimum fertilization (OPT) and application of 50% P₂O₅ increase based on OPT (OPT+P)) were selected in a 11-years fertilization filed experiment. Total P, Olsen-P and P loss in runoff were measured, and the relation of soil P and P balance were analyzed, agricultural threshold of Olsen-P content was also calculated by Mitscherlich model. The time needed for soil Olsen-P in 11th year to fall to the environmental threshold was further investigated.【Result】The total P under CK treatment decreased with a 0.0251 g·kg ^-1 decreasing rate year by year, while total P and Olsen-P content under fertilization treatment showed a fluctuation increment with planting time. Compared with OPT treatment, OPT+P treatment had a significant increasing rate for both total P and Olsen-P content. The soil P deficit amount was 10.2 kg·hm ^-2·a ^-1 under CK treatment, and soil P surplus amount was 41.3-69.2 kg·hm ^-2·a ^-1 for fertilizer treatment, which accounted for 31.9%-35.6% of P application rate, with 67.5% higher in OPT+P treatment than that in OPT treatment. There was significant correlation between the change of soil P and P balance (P<0.01). With 100 kg P·hm ^-2 cumulative surplus amount, soil total P increased by 0.06 g·kg ^-1 and 0.09 g·kg ^-1, and Olsen-P increased by 11.0 mg·kg ^-1 and 9.1 mg·kg ^-1 under OPT and OPT treatment, respectively. With 100 kg P·hm ^-2 cumulative deficit amount in CK treatment, total P decreased by 0.32 g·kg ^-1. Mitscherlich model fitted well the response between cane yield and Olsen-P content in lateritic red soil (P<0.01). The agricultural threshold of soil Olsen-P was 12.1 mg·kg ^-1. Fertilization significantly increased the P loss amount in surface runoff, and the amount of P loss under OPT + P treatment was higher than that in OPT treatment. The amount of P loss amount in surface runoff was positively correlated with Olsen-P content in soil. The time for Olsen-P decreased to the environmental threshold based on the relationship between the change of soil P, and P balance were 12 years and 16 years for OPT treatment and OPT+P treatment. 【Conclusion】In the latosolic red soil of south China, fertilization, especially with excessive P application, not only increased the soil P accumulation, but also increased the risk of P loss from surface runoff. P treated by OPT and applied every one year from the 2-3 year of planting could keep soil P in the range of agricultural threshold and environmental threshold based on the basic P content in this experiment.