【作者】 徐爱国；

【导师】 张维理；

【作者基本信息】 中国农业科学院 ， 植物营养学， 2009， 博士

【摘要】 以往的研究表明,农田地表径流引起的氮磷流失是地表水水体富营养化的主要原因。太湖水网地区雨量丰富,地势低平,农田排水沟渠密集,持续降雨条件下旱作农田也会发生渗漏,同样引起农田氮磷流失。流失方式不同、农田利用类型不同、施肥强度不同、土壤氮磷含量不同,农田氮磷流失特征与流失量都会有所不同。至目前为止,在太湖流域关于不同流失方式下,不同农田利用类型的氮磷流失特征的研究尚为缺少。本研究在浙江嘉兴、上海青浦区与松江区选择了12个实验点位,包括稻田、种植年限短菜地、种植年限长菜地3种农田类型,利用原位模拟降雨实验,在三种农田类型上,研究了渗漏与地表径流两种流失方式下的氮磷流失特征及流失量。主要结论如下:1. 12个点位的原位模拟降雨实验,降雨量为80mm条件下,旱作菜地农田地表径流水量和渗漏水量差异不显著,分别为降雨量的57.8%、64.6%;稻田地表径流水量为降雨量的89%,高于菜地径流水量。菜地农田实验点位上,地表径流方式下降雨结束后产流延续时间极显著低于渗漏方式,仅为渗漏方式的1/18,两者分别为1.7min、25.2min;而地表径流方式下产流速率是渗漏方式的3倍,两者平均分别为96.2ml/min、31.3 ml/min。2.三种农田类型上,地表径流方式下农田总氮流失量差异不显著,种植年限长菜地、种植年限短菜地、稻田总氮流失量平均分别为6.78、4.02和1.28 kg/hm²,总氮流失浓度平均分别为12.63、8.25、1.71 mg/L;渗漏方式下种植年限长菜地和种植年限短菜地总氮流失量差异也不显著,平均分别为78.83、35.62 kg/hm²,总氮流失浓度平均分别为164.32、56.32 mg/L。渗漏方式下硝态氮是流失总氮的主体,占总氮的70%以上;地表径流方式下水溶性氮为流失总氮的主体,平均占总氮的65%以上。在旱作菜地农田的8个点位上,渗漏方式下总氮流失量显著高于地表径流方式,渗漏和地表径流方式总氮流失量分别为32.5、5.4 kg/hm²。3.三种农田类型上,地表径流方式下农田总磷流失量差异不显著,种植年限长菜地、种植年限短菜地、稻田平均总磷流失量分别为0.93、0.36、0.13kg/ hm²,总磷流失浓度平均分别为1.83、0.78、0.17 mg/L;渗漏方式下种植年限长菜地和种植年限短菜地总磷流失量差异也不显著,平均分别为2.06、0.23kg/ hm²,种植年限长菜地总磷流失浓度显著高于种植年限短菜地,平均分别为3.09、0.38 mg/L。渗漏方式与地表径流方式下水溶性磷均占流失总磷50%以上,水溶性磷为渗漏方式和地表径流方式磷素流失的主体。渗漏方式与地表径流方式总磷流失量无显著差异,在菜地农田的7个点位上,渗漏方式与地表径流方式总磷流失量分别为0.62、0.71 kg/hm²。4.农田0⁵cm、0～20cm土壤硝态氮含量分别为31.2～472.9和33.2～171.1mg/kg时,地表径流方式下,农田0～5cm、0～20cm土壤硝态氮含量与径流液总氮、水溶性氮、硝态氮、铵态氮和水溶性有机氮流失量、流失浓度相关不显著。渗漏实验则显示:农田0-5cm、0～20cm土层硝态氮含量与渗漏液总氮、水溶性氮、硝态氮的流失量、流失浓度呈极显著正相关。0～20cm是氮素渗漏流失的关键层次,该土层NO₃^--N含量是渗漏氮素流失的主要影响因素,0～5cm土壤硝态氮含量对氮素渗漏流失的影响大于其他土层。5.农田0⁵cm、0～20cm土壤Olsen-P含量分别为3.9～59.6和2.4～45.0mg/kg时,地表径流方式下,土壤Olsen-P含量与径流液总磷、水溶性磷、水溶性无机磷、水溶性有机磷流失量、流失浓度相关不显著,与颗粒态磷流失量相关不显著。渗漏实验显示:0～5cm土壤Olsen-P含量低于20mg/kg时,渗漏液总磷浓度全部低于0.53 mg/L,流失量低于0.46 kg/hm²,土壤Olsen-P含量高于20mg/kg时,渗漏液中总磷浓度和流失量明显增高,总磷浓度均高于1.57 mg/L,最高达8.50 mg/L,流失量可达6.61 kg/hm²。本项研究结果显示,菜地农田在80mm降雨条件下,对氮流失而言,渗漏大于地表径流,即:在持续阴雨条件下,氮的流失风险大;而对磷流失而言,渗漏方式与地表径流方式区别不大。在菜地农田上,随农田表层0～5cm和0～20cm土壤硝态氮含量,0～5cm的土壤Olsen-P含量的增加,氮磷的地表径流流失或渗漏流失均有明显增加的趋势,换言之,土壤0～5cm硝态氮与Olsen-P含量可较好反映农田氮磷流失潜力。更多还原

【Abstract】 It was reported that N and P losses from cropping land into water system were mainly through runoff. Taihu Lake region is located at lowland plain with abundant precipitation and water-nets consisting of rivers, channels, ditches and flooding crop land. Under consisting raining period, which occur in winter or summer season, N and P loss through leaching may take place. Generally, N and P losses are depending on the losing procedures （runoff or leakage）, the cropping lands （vegetable or rice field）, the fertilizer rates and the soil nutrient concentrations. However, few works were done on N and P losses through runoff, leakage, from different cropland in Taihu Lake region. In this study, 12 site-specific rainfall simulation experiments were done in order to understand N and P losses through runoff, leakage and from different cropland. These experiments distributed in Shanghai and Zhejiang province and covered three different cropping land, i.e. rice filed, long-term vegetable land and vegetable land for less than 5 years. The main results were as follows.Under 80mm rainfall simulation, the results of 12 experiments showed that the amount of runoff water and leakage water in vegetable land were not different apparently, with 57.8% and 64.6% to the rainfall respectively. The amount of runoff water in rice field is higher than that of vegetable land, with an average of 89% to the rainfall. For vegetable land, the outflow lasting time after rainfall stopping under runoff is only 1/18 of that under leakage, with 1.7 minutes and 25.2 minutes respectively, but the velocity of runoff is 3 times faster than that of leakage, with 96.2 ml/min and 31.3 ml/min respectively.There are no apparent differences of total N loss through runoff among long-term vegetable land, vegetable land for less than 5 years and rice field, with 6.78、3.36和1.28 kg/ha ,respectively, and with average of total P concentration of 12.63、8.25、1.71 mg/L, respectively. And there are also no apparent differences of total N loss through leakage between long-term vegetable land and vegetable land for less than 5 years, with 78.83, 35.62 kg/ha, respectively,and with average of total N concentration of 164.32、56.32 mg/L, respectively. Analysis showed that NO₃^--N is the main form in leakage water, accounting for more than 70% of the TN, but that total dissolved nitrogen was the main form in runoff water. for that of runoff. The On the 8 sites of vegetable lands, analysis showed that total N losses （32.5 kg/ha） from leakage is significantly higher than that （ 5.4 kg/ha）of runoff.There are no apparent differences of total P loss through runoff among long-term vegetable land, vegetable land for less than 5 years and rice filed, with 0.93, 0.36 and 0.13 kg/ha,respectively, and with average of total P concentration of 1.83、0.78、0.17 mg/L. There are also no apparent differences of total P loss through leakage between long-term vegetable land and vegetable land for less than 5 years, with an average 2.06 , 0.23 kg/ha, respectively, and with average of total P concentration of 3.09、0.38 mg/L, respectively. Total dissolved P losses of runoff and leakage are both higher than 50% of total P losses. On the 7 sites of vegetable land, analysis showed that there are no apparent differences of total P between through runoff and through leakage, with an average of 0.71、0.62 kg/ha, respectively.Results from the runoff experiments showed that when soil NO₃^--N content is 31.2～472.9 and 33.2～171.1 mg/kg, there are no significant correlations between the content of 0⁵ cm or 0～20 cm soil NO₃^--N and concetrations of total N, total dissolved N, NO₃^--N, NH4+-N, particulate N in runoff water. But results from the leakage experiments showed that there were significant correlations between the content of 0⁵ cm or 0～20 cm soil NO₃^--N content and the concetrations or amount of total N, total dissolved nitrogen, and NO₃^--N. The depth of 0-20 cm was the main depth for leakage, and soil NO₃^--N content of this soil depth is the main factor for N loss in leakage. The effect of 0⁵ cm soil NO₃^--N content was higher than other soil depth.Results from the runoff experiments showed that when Olsen-P content is 3.9～59.6 and 2.4～45.0 mg/kg, there are no significant correlations between the Olsen-P content of 0⁵ cm or 0～20 cm soil depth and concetrations of total P, total dissolved P, dissolved reactive P, dissolved organophosphorus and particulate P. And results from the runoff experiments showed that when soil Olsen-P content is lower than 20 mg/kg, the total P losses is lower than 0.8 kg/ha. Results from the leakage experiments showed that when the content of 0⁵ cm soil Olsen-P is lower than 20 mg/kg, the total P concetration is lower than 0.53 mg/L, and the total P losses is lower than 0.46 kg/ha, but when the content of 0⁵ cm soil Olsen-P is higher than that value, the total P concetration is more than 1.57 mg/L, the maximum is 8.50 mg/L, and the total P losses is up to 6.61 kg/ha.Results from experiments showed that N loss through leakage was higher than through runoff on vegetable farmland under 80 mm rainfall simulation. That mean the risk of N loss was high under consisting raining condition. There were few difference between P loss through runoff and through leakage. With increase of NO₃^--N content of 0⁵ cm and 0²0 cm soil depth and increase of Olsen-P content of 0⁵ cm soil depth, N and P loss were obviously enhanced both through runoff and leakage. That mean NO₃^--N and Olsen-P content of 0⁵ cm could be well reflected the potential of N and P loss .更多还原