Response of soil Olsen-P to P budget under different long-term fertilization treatments in a fluvo-aquic soil

The concentration of soil Olsen-P is rapidly increasing in many parts of China, where P budget(P input minus P output) is the main factor influencing soil Olsen-P. Understanding the relationship between soil Olsen-P and P budget is useful in estimating soil Olsen-P content and conducting P management strategies. To address this, a long-term experiment(1991–2011) was performed on a fluvo-aquic soil in Beijing, China, where seven fertilization treatments were used to study the response of soil Olsen-P to P budget. The results showed that the relationship between the decrease in soil Olsen-P and P deficit could be simulated by a simple linear model. In treatments without P fertilization(CK, N, and NK), soil Olsen-P decreased by 2.4, 1.9, and 1.4 mg kg^(–1) for every 100 kg ha^(–1) of P deficit, respectively. Under conditions of P addition, the relationship between the increase in soil Olsen-P and P surplus could be divided into two stages. When P surplus was lower than the range of 729–884 kg ha^(–1), soil Olsen-P fluctuated over the course of the experimental period with chemical fertilizers(NP and NPK), and increased by 5.0 and 2.0 mg kg^(–1), respectively, when treated with chemical fertilizers combined with manure(NPKM and 1.5 NPKM) for every 100 kg ha^(–1) of P surplus. When P surplus was higher than the range of 729–884 kg ha^(–1), soil Olsen-P increased by 49.0 and 37.0 mg kg^(–1) in NPKM and 1.5 NPKM treatments, respectively, for every 100 kg ha^(–1) P surplus. The relationship between the increase in soil Olsen-P and P surplus could be simulated by two-segment linear models. The cumulative P budget at the turning point was defined as the "storage threshold" of a fluvo-aquic soil in Beijing, and the storage thresholds under NPKM and 1.5 NPKM were 729 and 884 kg ha^(–1)P for more adsorption sites. According to the critical soil P values(CPVs) and the relationship between soil Olsen-P and P budget, the quantity of P fertilizers for winter wheat could be increased and that of summer maize could be decreased based on the results of treatments in chemical fertilization. Additionally, when chemical fertilizers are combined with manures(NPKM and 1.5 NPKM), it could take approximately 9–11 years for soil Olsen-P to decrease to the critical soil P values of crops grown in the absence of P fertilizer.

长期不同施肥红壤磷素变化及其对产量的影响

【目的】定量长期不同施肥红壤磷素的演变特征,研究红壤磷素变化对生产力的影响,为红壤地区磷素管理提供理论依据。【方法】利用持续26年的红壤旱地长期定位试验平台(1991—2016年),比较长期不施磷肥(CK、N、NK)、施用化学磷肥(PK、NP、NPK)、化肥配合秸秆还田(NPKS)和化肥配施有机肥及有机肥(1.5NPKM、NPKM、M)土壤Olsen-P和全磷含量变化,分析土壤磷素对磷盈亏量的响应,采用不同模型拟合作物产量对有效磷的响应曲线,计算土壤有效磷农学阈值。【结果】长期施用磷肥显著提高了土壤全磷和有效磷含量,提升了土壤磷素活化系数(PAC)。化肥配施有机肥及有机肥处理(1.5NPKM、NPKM、M)的PAC高于化肥配合秸秆还田(NPKS)和施用化学磷肥(PK、NP、NPK)。红壤地区土壤全磷和有效磷变化量与土壤磷盈亏量呈正相关关系(P<0.01),土壤每累积盈余100 kg P·hm-2,土壤Olsen-P含量上升3.00—5.22 mg·kg-1,全磷上升0.02—0.06 g·kg-1。土壤每累积亏缺磷100 kg P·hm-2,不施磷肥处理(CK、N、NK)土壤Olsen-P分别下降1.85、0.40、1.76 mg·kg-1。化肥配施有机肥及有机肥处理(1.5NPKM、NPKM、M)的小麦和玉米产量显著高于化肥配合秸秆还田(NPKS)以及施用化学磷肥(PK、NP、NPK),显著高于不施磷肥(CK、NK、N)。化肥配施有机肥及有机肥处理(1.5NPKM、NPKM、M)的产量可持续指数也高于其他处理。3种模型(线性-线性模型、线性-平台模型和米切里西方程)均能较好地拟合作物产量与红壤有效磷含量的响应关系(P<0.01)。在红壤地区推荐使用拟合度较好的线性-线性模型,其计算出小麦和玉米的土壤Olsen-P农学阈值分别为13.5和23.4 mg·kg-1。【结论】在南方红壤地区,化肥配施有机肥更有利于磷素累积和提升磷素有效性。化肥配施有机肥作物产量显著高于其他处理,且稳产性好。线性-线性模型可用于计算红壤地区有效磷的农学阈值。生产上应该根据土壤有效磷含量及其农学阈值调整磷肥施用量。

长期施肥紫色土有效磷变化及其对稻麦轮作产量的影响

【目的】通过总结分析长期施肥处理下紫色土稻麦轮作土壤有效磷的变化特征,以及土壤磷素变化对作物产量的影响,为紫色土稻麦轮作磷素管理提供理论依据。【方法】依托国家肥力监测网紫色土肥力监测试验站27年的稻麦轮作定位试验,选取10种不同施肥处理:CK处理(只种作物不施肥);N、NP、NK、PK、NPK为不同氮(N)、磷(P)、钾(K)化肥配施处理;M、NPKS、NPKM、1.5NPK+M为有机肥(M)、秸秆还田(S)及其与化肥配施处理。试验数据涵盖1991—2018年,测定不同施肥处理下土壤有效磷含量和作物产量,计算100 kg籽粒磷素吸收量和磷肥利用率,分析土壤磷素变化对累积磷盈亏的响应,采用不同模型计算土壤磷素农学阈值。【结果】长期施用磷肥能够显著提高土壤有效磷含量,各施磷处理有效磷年均增量为0.80—2.32 mg·kg^(-1);而不施磷处理CK、N、NK和单施有机肥处理M的土壤有效磷含量则逐年下降至平稳状态。不施磷处理土壤磷素一直处于亏缺状态,施磷各处理27年后土壤累积磷盈余量为244.8—698.2 kg P·hm^(-1),其中1.5NPK+M处理累积磷盈余量最高;施磷处理土壤累积盈余量与土壤Olsen-P增量呈显著线性相关,土壤每盈余磷100 kg·hm^(-1),土壤有效磷含量提高4.27—6.5 mg·kg^(-1)。磷肥施用能显著提升稻麦轮作系统作物产量和吸磷量,100 kg水稻籽粒需磷量为0.17—0.41 kg,100 kg小麦籽粒需磷量为0.25—0.57 kg;试验各处理的磷肥利用率为10.3%—39.7%;4种模型(线性-平台模型、双直线模型、BoxLucas模型和米切里西模型)均能较好地拟合作物产量与紫色土有效磷含量的响应关系,其中双直线模型的拟合度最好,其计算的水稻和小麦的土壤有效磷农学阈值分别为13.28和9.93 mg·kg^(-1)。【结论】在紫色土水稻-小麦轮作体系中,合理施用磷肥能显著提高作物吸磷量、产量以及土壤有效磷含量。推荐双直线模型用于计算紫色土稻麦轮作体系下土壤有效磷的农学阈值,生产上应根据土壤有效磷含量及其农学阈值调整磷肥施用量。