【作者】 臧玲；

【导师】 田光明； 梁新强；

【作者基本信息】 浙江大学 ， 环境工程， 2011， 博士

【摘要】 鉴于胶体在污染物分布特征及元素循环中的作用以及富营养化的日益加剧,胶体态磷以其流失量大,生物利用性高等特点逐渐成为面源磷污染研究的新热点。本研究以浙江嘉兴地区典型水稻土为对象,通过长期田间小区试验、径流模拟试验、室内淋溶试验、静态批处理试验研究了不同磷饱和度的土壤中胶体细颗粒物及胶体态磷的分布特征；考察了化肥和有机肥施用的胶体态磷流失源强及其径流、淋溶迁移特征,为全面认识磷素流失和有效评价磷素流失的环境风险提供了依据。通过田间小区试验研究了施肥对不同粒径土壤磷素的分布与流失贡献的影响；探讨了稻田胶体态磷素流失特征及其施肥效应；研究了猪粪有机肥释放的水分散性胶体对不同类型土壤胶体态磷的活化迁移影响；通过模拟试验分析了稻田磷素流失的形态及主控因子。主要研究结果如下：1.土壤各粒级组分在理化性质上有显著差异,不同组分吸附磷的环境学归宿也显著不同。以嘉兴典型水稻土不同施肥处理的试验小区土壤为对象,研究了不同粒级组分(粘粒<2μm,粉粒2-20μm,细砂粒20-200μm,粗砂粒200-2000μmm)中磷的分布及等温吸附曲线,得出了各组分对土壤吸附磷的贡献率,并采用SPSS软件对影响吸附的因子进行了回归分析。结果表明,各粒级组分中总磷的分布为：粘粒(52.84%±0.93%)>细砂粒(24.85%±2.47%)>粗砂粒(16.72%±2.69%)>粉粒(9.09%±1.48%)；速效磷占总磷比例随粒径的增大而减小,粘粒组分中磷含量最高,约50%以上的总磷和69%以上的速效磷吸附在粘粒组分上。不同粒级组分对磷的吸附量在各处理土壤中均表现为：粘粒>粉粒>粗砂粒>细砂粒,粘粒对磷的最大吸附量分别是粉粒的1.30倍、细砂粒的1.61倍和粗砂粒的1.40倍。粒级、有机质、有效铁均对土壤吸附磷有较大影响。2.通过田间小区试验研究了土壤的磷素状况与环境磷素间的相互关系,考察了土壤胶体态磷和溶解态磷在土壤剖面的迁移情况,建立了磷素淋失的评价指标。结果表明,土壤全磷(TP)、胶体态磷(Colloidal P)、水溶性磷(Dissolved P)、速效磷(Olsen P)在土壤剖面上均存在较大的变幅,磷素在土壤中积累量均随施肥量增大而增大,有机肥处理中磷素向下迁移的趋势更加明显。通过分段线性模型分析了土壤剖面上胶体态磷与DPSox、水溶性磷与DPSox、Olsen-P与DPSox的关系,结果表明,水溶性磷和Olsen-P分别在DPSox为9%和12%时存在明显的突变,超过该突变点,则水溶性磷和Olsen-P明显有向下迁移的趋势。而分段线性模型中DPSox与胶体态磷的没有发现明显的突变点,但二者之间存在极显著的线性相关。由相关性分析可见,虽然土壤剖面上的胶体浓度与土壤pH(正相关)及离子强度(负相关)呈显著的线性相关,但土壤pH及离子强度与胶体态磷浓度相关性不显著。尽管底层土壤中土壤磷饱和度较低,但底层土壤中胶体态磷含量仍然较高,这可能是由于受底层土壤中较高pH及较低的离子强度的影响,大量土壤胶体仍可从土壤固相释放出来,而胶体态磷也被土壤胶体携带释放。可见,土壤磷饱和度对胶体态磷迁移有重要的影响。而高的pH、低的离子强度对底层土壤胶体和胶体态磷的释放影响也不易忽略。3.通过模拟降雨试验研究了土壤胶体态磷与溶解磷的径流流失特征,并分析了土壤中初始磷(非当季施肥入的磷肥)和新施入的肥料磷对径流流失磷的贡献。结果表明：在所有处理中,施肥量、土壤初始磷与总磷流失负荷均存在显著正相关关系；在未施入新肥料磷的处理中,有机肥处理的土壤中流失的磷明显高于无机肥处理土壤,而在实验前24h分别施入有机和无机磷肥后,无机肥土壤中流失的磷量显著增加并超过有机肥土壤中各形态磷素的流失磷量。其中胶体态磷(0.1-1μm)占流失总磷42-62%,而总溶解磷(TDP) (<0.45μm)占总磷的60%以上,是流失磷的主要形态。与不施无机肥的对照组相比,新施入无机肥的处理中,胶体态磷和溶解态磷流失量均显著增加,二者占总磷的比例也显著增加；与不施有机肥的对照组相比,新施入有机肥的处理中,胶体态磷和溶解态磷流失均有所增加,但占流失总磷的比例变化略有减小。此外,线性回归分析表明了土壤新施入的磷肥是流失负荷中磷素的主要来源。4.利用土柱模拟试验,在饱和流条件下研究了外源水分散性有机肥胶体、pH、离子强度对水稻土中胶体态磷素活化迁移(Colloid Facilitated Transport)的影响。结果表明,与去离子水作为淋入液的处理相比,在初始阶段,有机肥胶体的输入明显抑制了溶解态磷的迁移,溶解态磷的淋出浓度远远小于去离子水处理中的浓度,待土柱中吸附点位被饱和后,淋滤液溶解态磷的淋出浓度迅速增加,并最终接受有机肥胶体悬浮液中磷的浓度；而淋出液中胶体态磷的变化则不同,随有机肥胶体的输入不断增加,淋滤液胶体态磷浓度也不断增加。整个实验过程中,淋滤液胶体态磷浓度始终较去离子水处理中的淋滤液胶体态磷浓度高,且与胶体态铁存在着显著的线性关系。铁化合物可能是胶体态磷运输的主要载体。此外,研究也表明,高的DPSox不仅会促进溶解态磷的迁移,同时会促进胶体态磷的垂直迁移；流入液中pH的增加、离子强度的降低有利于土壤中胶体态磷的释放。5.通过田间小区试验和静态培养实验,研究了不同施磷水平对嘉兴水稻土中磷素积累、土壤磷饱和度以及流失水体中磷的影响。结果表明,DPSM3和DPSox两种土壤饱和度计算方法、几种有效磷之间具有较好的相关性。两种饱和度对于供试土壤饱和度的预测都是可行的。DPSox作为土壤磷素水平和固磷能力的综合指标,代表了土壤吸附磷素和释放能力的强弱,在研究土壤磷素径流流失对地表水质的潜在影响时,不仅能很好地表征土壤磷素向环境迁移流失的潜能,也能有效的表征土壤剖面中磷素的淋失和土壤胶体态磷的流失潜能。更多还原

【Abstract】 Considering the role of the colloid in the pollution distribution and elements cycling, the presence of colloids in runoff is very important to understand the colloidal phosphorus (P) transport process. Therefore, the transport of colloidal P in soils has received considerable attention in non-point pollutions in recent years, because colloid-facilitated phosphorus has been proved as a significant contributor to eutrophication. The overall objectives of this research has been to assess the effect of drying-flooding cultivation on the potential of P distribution, and to discover the characteristics of P release to water bodies from the paddy field in field scale. In the present work, a long-term experiment was set-up in 2005. From the experiment, the distribution of P in soil profile of the paddy field can be monitored, including the mobilization of colloidal phosphorus, dissolved P and Olsen P from topsoils to subsoil. Batch experiments were also conducted to investigate the equilibrium sorption isotherms for P onto particle-sized fractions. Besides, to determine the effect of water dispersible colloids derived from swine manure on the leaching of phosphorus from paddy soils, leaching experiment in saturated-flow columns packed with aggregate-sand mixture materials were investigated using manured soil and unmanured soil. The detailed results are as follows.(1) Batch experiments were conducted to investigate the equilibrium sorption isotherms for P onto particle-sized fractions (<2μm,2～20μm,20～200μm, and 200～2000μm) which were named clay, silt, fine sand and coarse sand according to international system respectively) derived from a series of long-term paddy field of Jiaxing in China. The results showed that the ratio of total P (TP) in each fraction to total P in the bulk soil followed by the order:clay (52.84%±0.93%)> fine sand (24.85%±2.47%)> coarse sand (16.72%±2.69%)> silt (9.09%±1.48%). All of he percent of TP and Olsen P adsorbed by clay of the three soils exceeded 50% and 69% of the soil, respectively. The adsorption of P by these fractions could be described by the Langmiur equation and Frendlich equation. The maximum adsorption of P (Cm) and distribution coefficient (Kd) decreased with the following order:clay>silt>coarse sand>fine sand in the three soils. By choosing particle size, organic matter, iron oxide and equilibrium pH value as parameters, multivariate statistical analysis in SPSS were employed to build the regression model of adsorption of P and to evaluate the effect of these factors on it. The results showed that particle-size fraction, organic matter have much more significantly effect on the adsorption of P than available iron.(2) Application of P with animal manure and fertilizer in amounts exceeding removal with crops leads to accumulation of P in soil, making them potential long-term diffuse sources of P loss to water. The impact of a range of manuring and fertilization practices on the TP, Olsen P, distribution of dissolve phosphorus and colloidal P, and degree of phosphorus saturation (DPS) of soil were investigated in field study. In the present work, the overall comparison of all sites with a wide range of DPS has been investigated. The results indicated that DPS was an important factor in controlling the concentration of dissolved P and colloidal P in soil. The change points at 9% and 12% DPS were noted by using a split-line model, above which Olsen P (10.8 mg P kg-1) and dissolved P (3.1 mg P kg-1) in soil profile began to rapidly increase and potentially mobile downward. Therefore, it is supposed that the leaching of dissolved P can not be neglected as a widespread environmental problem. Compare with dissolved P, colloidal P was the dominant fraction of P in water-dispersible colloid suspension of the soil profile. The significant decrease of ionic strength and the increased pH value from topsoils to subsoils can explain the high release of soil colloid. And the soil colloid was the important carriers of collidal P. Therefore, colloidal P was also high in the subsoil despite DPS was low in subsoils. Overall, the high DPS induced by manuring and fertilization was the main factor of the transportation of colloidal P. But the effect of pH value, EC on the release of colloidal P in the soil also can not be overlooked.(3) The objective of this study was to test whether new application of fertilization 24 h before an intense rainstorm or the intial P in the soil is the main source of P loss in runoff water. A rainfall simulation study compared TP, colloidal P, total dissolve phosphorus (TDP), and total particulate phosphorus (TPP) concentrations and losses in runoff water after swine manure and inorganic fertilizer were broadcast. The results showed that P loss increased with applications fertilizer or manure and initial soil P of the soil, with most occurring as TDP accounted for more than 60% of total P in the fraction<0.45μm. Colloidal P also account for 42-62% of total P in the fraction 0.1-1μm. In comparison with IP1-C and IP2-C of group 1 (without application of fertilizer), the concentration of Colloidal P, TDP were much higher after using inorganic fertilizer in IP1 and IP2 treatments of group 2, respectively. And the ratio of colloidal P and TDP to TP also increased relatively in these treatments, respectively. Howerer, in comparison with OP1-C and OP2-C of group 1 (without application of manure), the concentration of Colloidal P, TDP were also increased after the application of maure in OP1 and OP2 treatments of group 2, respectively. But the ratio of colloidal P and TDP to TP decreased relatively in these treatments, respectively. The linear regression equation analysis demonstrated that new applied fertilizer and manure were the main source of runoff P rather than intial P in soils.(4) To investigate the effect of water-dispersible colloids derived from swine manure on the potential risks of P, migration behavior of P in saturated-flow columns were compared in the presence and absence of water-dispersible colloids of manure in the inflow. It was found that total dissolved P (TDP) accounted for a majority of total P (65%-98%) in the effluent with deionized water treatments, while only accounted for 21%-45% to total P in the leachate in the manure colloid treatments. In manured soils, with the inflow of manure colloidal suspension, colloidal P in the effluent were 26.7 times higher than that of deionized water treatment (PM+W) and 1.9 times more than that of unamended soil treated with manure colloid (PO+M) in the end of the leaching experiment, respectively. Despite the initial reduction of TDP concentrations in the effluent with the presence of manure colloid, the TDP concentrations still increased smoothly and continued to transport with the effluent throughout the breakthrough experiment. This suggested that P sorption sites of the soil and the added manure colloid in the column were fastly saturated during initial stage of the experiment. The good linear correlation between colloidal P and colloidal Fe indicated that Fe hydroxides could be served as a main medium for the transportation of colloidal P. Moreover, colloidal P exhibited greater mobility under higher pH and lower ionic strength.(5) In a series of laboratory soil incubation experiments, the effect of applied phosphorus on soil test phosphorus, degree of phosphorus saturation (DPS), dissolve phosphorus in water were discussed. During the experiment, two types of degree of phosphorus saturation had an obvious increase as a result of applied phosphorus. In a series of laboratory soil, DPSox increased from 11.2% to 34%, while DPSM3 increased from 2.72% to 17.98%. In addition, in both of the surface runoff of P and mobilization of P in soil profile, the statistics analysis found significant relationships between the differient DPS and soil test phosphorus. Therefore, the DPS was a good indicator of environmental impacts of P loss potential from agricultural soils to waters.更多还原