【作者】 李学平；

【导师】 石孝均；

【作者基本信息】 西南大学 ， 农业资源利用， 2008， 博士

【摘要】 磷是植物生长必需的营养元素,但不合理过量施用时,又成为环境污染因子。为了实现作物高产和环境友好的双重目标,必须将土壤有效磷水平维持在既能满足作物高产需求又能降低磷流失风险的环境阈值之间。因此,研究农田土壤磷素的迁移动态、流失特征及影响因素,弄清土壤磷累积与作物高产及磷素流失之间的关系,探讨土壤磷素环境风险评估的方法和指标,对于降低磷素流失、控制农业磷的面源污染、制定磷肥优化管理措施具有重要意义。国内外对旱地磷素流失进行了大量研究,但对稻田磷素流失研究较少,关于水旱轮作、长期不同施肥条件下紫色土稻田磷素的迁移流失及环境效应未见报道。水旱轮作由于干湿交替的水分管理,磷素在土-水中的迁移和淋失必定不同于旱地和湿地;长期不同的施肥方式必然导致土壤的物理、化学和生物学特性变化,土壤对磷的吸附固定能力发生变化,进而影响磷素在土-水界面的迁移转化。为此,本文采用田间长期定位试验与短期试验、渗漏池试验与室内模拟试验相结合的方法,通过15年水早轮作长期定位试验研究了长期不同施肥处理紫色土磷素累积状况及其在土壤剖面和水体中的迁移特征,探讨了长期施肥条件下紫色土对磷的吸附解吸特性及其影响因素,为紫色土水旱轮作稻田磷肥的优化管理和控制稻田磷素流失提供依据。采用单排单灌的田间试验,在水稻生长季节研究不同磷肥用量和不同有机肥品种对土-水磷素迁移流失的影响,并对其可能产生的环境风险进行分析,为紫色土区稻田水肥的优化管理提供理论依据。利用原状土壤渗漏池,研究水旱轮作条件下不同类型紫色土磷素渗漏淋失特征及其对地下水体环境质量的影响。通过土壤测试和模拟试验分析了土壤有效磷水平与磷素淋失之间的关系,探讨了紫色土磷素环境阈值,为紫色土区农田磷素环境风险评价提供依据。15年长期定位试验结果表明,紫色土磷素盈亏量（x）与土壤耕层Olsen-P的增减（y）呈极显著正相关:y=0.0394x+5.923（r=0.944**,n=11）,紫色土每积累100kg·hm^-2的磷素,土壤Olsen-P提高3.94mg·kg^-1。水旱轮作每年施用磷肥120kg·hm^-2（P₂O₅,常规施磷水平）,磷素盈余53.9kg·hm^-2,提高了土壤速效磷含量;有机肥与化学磷肥配合施用能加快土壤速效磷的积累。长期施用在表层土壤的磷肥有向下移动的趋势,施磷处理20～60cm土壤速效磷含量比不施磷处理高,在水旱轮作条件下磷己迁移至60cm土层,增加了磷渗漏淋失的风险。磷在土壤剖面中的迁移能力受磷肥用量、有机肥种类和种植方式的影响,向下迁移的磷数量随着磷肥施用量的增加而增加;在等养分情况下,配施猪粪比配施秸秆更容易导致磷的迁移;稻油轮作土壤剖面中有效磷含量比稻麦轮作高1.25～6.5mg·kg^-1,稻油轮作体系磷素的迁移能力比稻麦轮作强,磷素流失的风险增大。从长期定位试验水稻季土-水磷素动态变化看,土壤磷向水体的迁移主要发生在淹水种稻后的前10天,所有处理水稻移栽后l0天田面水磷含量较高,之后急剧下降;施肥35天后各处理田面水磷含量低(在0.1 mg·L^-1左右波动)并趋于稳定。在水稻移栽前10天应禁止稻田排水,在水稻移栽后30天内最好不要进行稻田排水,采取浅水移栽的稻田水分管理策略可以减少磷素流失。不同施肥处理之间,田面水含磷量在水稻移栽的前30天差异较大,以高量施磷处理1.5NPK+M的TP浓度最高;其次为配施猪粪尿的处理,不施磷的处理最低;稻—油轮作田面水TP浓度高于稻—麦轮作;配施猪粪提高了土壤磷素向水体的迁移和流失。土壤易解吸磷（RDP）、活性磷（藻类有效磷）、磷零吸持平衡浓度（EPC₀）、最大吸磷量（Qm）、磷吸持指数（SI）等参数是评价水-土界面磷迁移能力的重要指标。长期倍量施磷（1.5NPK+M）以及化肥与猪粪配施（NPK+M）处理土壤磷素在土壤剖面和水体的迁移能力高于其它处理的机理在于这2个处理土壤的易解吸磷、活性磷、磷零点吸持平衡浓度、磷吸持饱和度最高,土壤对磷的固定能力和缓冲能力降低,提高了磷向环境迁移的风险。在水旱轮作条件下土壤对磷的固定能力与土壤活性铝、pH呈极显著正相关;与土壤有机质和土壤Olsen-P含量呈极显著负相关。不同磷肥用量和有机肥对稻田磷素迁移的试验结果表明,田面水磷浓度随着施磷水平的提高而提高,施肥24h后各处理田面水磷素浓度最高,TP浓度在0.928～3.824mg·L^-1之间。前10d田面水总磷浓度高,平均浓度为0.869mg·L^-1,前30d田面水含磷波动大,各处理TP平均浓度在0.258～1.433mg·L_-1之间,这期间的田间排水会对周围的水体环境造成污染。径流水各形态磷的浓度随着降雨强度和磷肥用量的增加而增加,有机无机肥配施的处理稻田土壤磷素向水体释放磷的持续时间更长,施用牛粪比施用秸秆更容易导致磷素的流失。稻田磷素流失总负荷随着磷肥用量的增加而增加,流失磷量最低和最高分别为0.52kg·hm^-2和3.20kg·hm^-2。稻田磷素流失负荷不仅与施用磷肥的量有关,而且与有机肥的种类有关,配施秸秆能减少土壤磷的流失。磷素渗漏淋失试验结果表明,水旱轮作条件下,无论是旱季还是稻季在3种紫色土上磷素渗漏淋失量都随着施磷量的增加而增加。磷素的渗漏淋失与土壤性质关系较大,旱季和稻季在3种类型紫色土上磷素渗漏淋失浓度和渗漏淋失负荷为中性紫色土＞钙质紫色土＞酸性紫色土;在100cm土层稻季的磷渗漏量在75.4～158.2g·hm^-2之间,麦季在12.0～25.2g·hm^-2之间,稻季是旱季的5～8倍。在水稻生长季不同土层渗漏水中总磷浓度随着水稻生育期的延长呈下降趋势,前期磷素渗漏淋失量高且变幅较大,在施磷后第5d各土层渗漏水中磷素浓度最高;施肥60d后稻田磷素渗漏淋失负荷显著降低,不同土层和各处理趋于一致。渗漏水中磷素浓度随土壤剖面深度的增加而降低,磷在土壤剖面中的迁移能力弱,整个水稻生育期磷渗漏量较低,最高磷渗漏量为0.262kg·hm^-2。但是,在水稻移栽初期中性和钙质紫色土耕层磷肥有明显向下移动的趋势,表现为60cm土层渗漏水中磷浓度高,且施磷处理显著高于不施磷处理,淹水种稻初期是磷向下迁移最强和淋失量最高的时期,控制基肥中磷肥的施用是减少淹水初期磷素向下迁移的重要措施。水旱轮作条件下,为了降低灌水种稻初期由于水分的剧烈移动而导致的磷素渗漏淋失,可以在水稻移栽成活后施用磷肥。利用具有不同富磷梯度的三种紫色土模拟稻田和旱地研究了Olsen-P与CaCl₂-P之间的关系,探讨了紫色土磷素环境阈值。结果表明,无论是稻田或旱地生境,三种紫色土Olsen-P与CaCl₂-P之间都存在一个‘转折点”,当土壤Olsen-P含量低于转折点时,随着Olsen-P含量的增加,浸提液中CaCl₂-P含量上升很小,但当土壤Olsen-P含量高于转折点时,CaCl₂-P含量急剧增加,这个点就是旱地磷素淋失临界值或稻田土壤磷的环境警戒值。紫色土旱地磷素环境敏感淋失临界值在酸性、中性和钙质上分别为（Olsen-P含量）67.2、85.8和113.8mg·kg^-1,与此对应的CaCl₂-P含量分别为2.5、1.6、2.3mg·kg^-1。三种紫色土磷的淋失临界值差别较大,受土壤性质和对磷的吸附解吸特性影响,钙质紫色土对磷的吸附固定能力强,最大吸附量(Qm为769.2mg·kg^-1)比中性和酸性土高20%和60%。稻田土壤磷素的环境警戒值在酸性、中性和钙质紫色土上Olsen-P含量分别为49.2、77.9和92.1mg·kg^-1,相应的CaCl₂-P含量分别为2.0、1.4、1.2mg·kg^-1。三种紫色土在淹水还原条件下土壤磷环境敏感临界值比旱地低,淹水还原条件提高了紫色土磷向水体释放的风险。根据土壤测试而获得的紫色土最大吸磷量（Qm）、磷吸持指数（SI）、易解吸磷（RDP）、活性磷（藻类有效磷）和磷零吸持平衡浓度（EPC₀）等参数,可以作为表征水-土界面磷迁移能力的指标。磷吸持指数（SI）可以作为Qm的替代指标来评价磷流失风险,将SI值35作为评价紫色土磷素流失的临界值,当低于此值时,紫色土固磷能力低,土壤磷素流失的风险就大;反之亦然。可以利用这些指标对紫色土区域土壤磷环境风险进行评价,并确定区域磷肥的最佳管理策略。更多还原

【Abstract】 Phosphorus（P）management for sustainable crop production and water quality protection is in great concern worldwide.The mobility of P in terrestrial ecosystems poses water quality concern because of its contribution to eutrotrophication.Agricultural land uses are considered a leading source of P entering the rivers and lakes.Many factors,including soil properties（soil moisture,soil texture,mineralogical composition and P fractionation）,climate conditions（precipitation）,topographic features,and management practices（tillage,irrigation,fertilizer application,crop rotation）interact to influence the loss of P from cropland soils.As a consequence,the potential of P loss from cropland soils has high spatial and temporal variability.Many researches have been carried on in upland field about running off of phosphorus but few done in paddy field especially in rice-wheat rotation and long-term different fertilization in purple soil.For the different management of water in the alternation of drying and wetting,P transport and leaching running off must be different with upland field and wetland and the changing physics,chemistry and biologic biological trait of soil also influence the capability of soil absorbing and desorbing P as a result of long-term different fertilization which affect P transplant in soil-water interface eventually.Therefore,a study was carried out to address the following questions:1)Long-term fertilization effect on P accumulation in the purple soil subjected to a rice-wheat rotation system.2)Quantification of leaching loss of P during rice season using the lysimeter method.3)P fertilization effect on the P dynamic in the surface water of paddy field.4) Identification of the soil available P indices that can be used to predict P loss from paddy field.The results showed that Soil P balance（x）and soil Olsen-P change（y）was very significant positive correlations:y=O.0394x+5.923(r=0.944^**,n=11)in 15 years of long-term experiment.Soil Olsen-P content increased 3.94 mg·kg^-1when soil accumulated 100kgP·hm^-2.The P surplus was 53.9kg·hm^-2and content of soil available P increased when P application rate was 120 kg·hm^-2（P₂O₅,normal P application rate）.Organic manure combining with inorganic fertilizer increased accumulation of soil Olsen-P.P fertilizer moved downwards for long-term fertilizing P and soil available P was higher in treatment of fertilizing P than no P fertilization in 20～60cm.The risk of running off P increased as P moving to 60cm soil layer in rice-wheat rotation.P moving downwards was influenced by amount of P fertilizer,kind of organic manure and planting methods.The quantity of P moving downwards increased with the increase of P application rote.P moving was easier in soil of chemical P application combined with pig excrement than combined with straw on the condition of equivalent nutrient.Olsen-P content was higher 1.25～6.5mg·kg^-1 in rice-rape rotation than in rice-wheat rotation and P moving capability was stronger in rice-rape rotation than in rice-wheat rotation and the risk of P running off also increased in rice-rape rotation.P moving downwards water bodies mainly happened in first 10 days after paddy flooded water based on dynamic change of soil-water P in paddy season in long-term fertilization,P concentration of surface water higher in 10 days after paddy transplanted than other times and later dropping sharply.P content of surface water was low and tended towards stability after fertilization about 35 days(fluctuating in 0.1mol.L^-1).It was forbidden to drainage water in the first 10 days after paddy transplanted and better not to drainage water in 30 days after paddy transplanted.Paddy was transplanted in shallow water which could reduce P running off,TP concentration of surface water of every treatment differed greatly in 30 days after paddy transplanted and highest TP concentration was treatment of 1.5NPK+M,treatment of combining with pig excrement secondly and no P application lowest.TP concentration was higher in rice-rape rotation than rice-wheat rotation and combining with pig excrement increased soil P transporting and running off.RDP（Readily Desorbable P）,BAP（Bioavailable P）,EPC₀（The equilibrium P concentration at zero sorption）,Qm and SI（Sorption Index）were the important indexes of evaluating P moving ability of soil-water interface.The mechanism was that RDP,BAP,EPC₀ and DPSS（Degree of Sorption Saturation） of 1.SNPK+M and NPK+M（combing with pig excrement）treatments were the highest,soil firming and buffering capacity decreased,environmental risk of P running off increased,and so P moving ability was higher in soil profile and water bodies in 1.5NPK+M and NPK+M（combing with pig excrement） treatments than other treatments.Capacity of soil firming P present extremely remarkable positive correlation with soil active Al,pH,and present extremely remarkable negative correlation with soil organic manure and Olsen-P.The results showed that TP concentrations in the surface water increased as P application rate enhancing and reached the peak value at the range of 0.928～3.824 mg·L^-1after 24 h in all the treatments based on P running off in paddy field in using different P application rate and organic manure.TP concentration was high and the average concentration was 0.869mg·L^-1in 10d after fertilization. Fluctuation of TP concentration in surface water was drastic,the average TP concentration was 0.259～1.433 mg·L^-1,and drainage would pollute circumstance of water bodies during the first 30 days.The contents of different P forms in the runoff water increased with the increase of precipitation and P application rate.The time of soil releasing P would last longer in treatment of organic-inorganic fertilizer than others and P running off was easier in using cattle manure than straw.Total loading of P running off increased with the increase of P application rate and the highest and lowest loadings were 0.52kg·hm^-2and 3.20kg·hm^-2,respectively.The loading of P running off not only had relation with P application rate but also with kind of organic manure.Straw combining with inorganic fertilizer could decrease P running off.The results showed in lysimeter experiment that P concentration in leachate increased with P application rate in three kinds of soil in rice-wheat rotation,whether was in dry season or paddy season. Soil characteristic influenced greatly P leakage.Concentration and loading of P running off was neutral soil＞calcium soil＞acid soil in dry season and paddy season.Leakage amount of P was at the range of 75.4～158.2g·hm^-2on 100cm profile and leakage amount was 12.0～25.2 g·hm^-2in dry season,which in rice season was 5～8 times of dry season.P seepage appeared downtrend in the whole,and the leakage amount was high and the variation range was bigger in earlier stage.Concentration of P seepage was the highest in all soil layers 5 days after fertilization,remarkably decreased 60 days after fertilization,and reached coincidently at different soil profile with all treatments.P concentration of seepage water decreased with increasing the depth of soil profile and the ability of P moving was weak in soil profile.P leakage amount was low in the whole paddy season,and the highest value was 0.262kg·hm^-2.P fertilizer moved downward obviously in neutral and calcium soil in the initial stage of paddy transplanted and P concentration of seepage water was the highest in 60cm layer,also the treatment of P fertilization was higher obviously than no P fertilization.The strongest time of P movement was earlier stage of paddy transplanted and the good measure which decreased P moving downward was to control basal dressing.In rice-wheat rotation,the stage of P fertilization was in the time after paddy transplanted surviving in which decreasing P leakage because of violent movement of water bodies.The relationship between Olsen-P and CaCl₂-P was studied using soil of different rich P gradient, which simulated in three kinds of soil,to elucidate P critical point of purple soil.The results showed that there was a "change-point" between Olsen-P and CaCl₂-P in three kinds of soil whether in paddy field or upland field.CaCl₂-P content of extracting solution with the increase of Olsen-P content was very low at all times when soil Olsen-P content was lower than the change-point,but CaCl₂-P content increased sharp when soil Olsen-P content was higher than the change-point,and the change-point was called critical value of upland leakage or break-point of paddy field.The critical value of purple soil in upland field was 67.2,85.8 and 113.8mg·kg^-1in acid soil,neutral soil and calcium soil,respectively,and CaCl₂-P content corresponding was 2.5,1.6 and 2.3 mg·kg^-1.The critical value had great difference in three kinds of purple soil and was affected by soil character and characteristics of P-absorbtion and P-desorption.The capacity of absorbing P in calcium purple soil was strong and Qm of calcium soil(769.2 mg·kg^-1)was higher 20%and 60%than that of neutral and acid soil, respectively.The critical value of purple soil in paddy field was 49.2,77.9 and 92.1mg·kg^-1in acid soil,neutral soil and calcium soil,respectively,and CaCl₂-P content corresponding was 2.0,1.4 and 1.2mg·kg^-1.The sensitive critical value of purple soil P was lower in the condition of flooding reduction than in upland field, and flooding reduction increased the risk of P releasing into water circumstance.The indexes of Qm,SI,RDP,BAP,EPC₀,and so on were regard as to signify capacity of P moving in soil-water interface.SI could substitute Qm to evaluate environmental risk of P running off in purple soil and SI value 35 was regarded as a critical value.Soil had a capacity of soil firming P was weak below value 35 and the risk of P running off was high.And the same was on the contrary.Using these indexes can help evaluate the environmental risk of P running off in purple soil region and establish the best management strategy.更多还原