节点文献

湘江与洞庭湖水体氮素时空变化特征及湘江水体中氮浓度预测方法研究

Studies on the Temporal and Spatial Variation Characteristics of Nitrogen in Xiang Jiang River and Dong Ting Lake and Prediction Method of Nitrogen Concentration in Xiang Jiang River

【作者】 周九州

【导师】 刘强; 荣湘民;

【作者基本信息】 湖南农业大学 , 植物营养学, 2010, 博士

【摘要】 湘江是湖南的母亲河,水体污染日趋严重引起社会的广泛关注,随着点源污染的有效控制和治理,农业非点源造成的氮污染成为水体污染的主要因素。本文在全面分析湘江与洞庭湖水体氮素时空变化特征的基础上开展水稻合理施氮研究,为防止湘江发生富营养化、抑制水体氮进一步恶化提供科学依据;同时,优选出湘江水体中氮浓度最佳预测模型,为今后在水环境规划管理和水污染综合防治过程中对湘江水质模拟预测提供一种新的方法和思路。研究主要内容与结论如下:对湘江、洞庭湖水环境因子(水温、pH值、溶解氧)与水体中总氮、氨氮、硝态氮的年度变异特征进行分析,研究结果表明:2006年4月-2007年3月,湘江、洞庭湖水体中水温和溶解氧均表现出显著的季节性变化特征、pH值年内变化不显著;湘江水体中氨氮、硝态氮、总氮浓度变化范围分别为0.03-1.76mg/L、0.02-1.68mg/L、1.11-7.62mg/L,洞庭湖水体中氨氮、硝态氮、总氮浓度变化范围分别为0.03-1.56mg/L、0.04-1.72mg/L、0.04-9.81mg/L;非点源污染对湘江、洞庭湖水体中总氮浓度影响比较大,年内峰值出现在农忙且降水较多的7月或8月;湘江熬洲、乔口以及洞庭湖的万子湖、目平湖等四个断面水体中总氮浓度与水温之间存在显著正相关;湘江熬洲、乔口、鹿角以及洞庭湖的万子湖、目平湖等五个断面水体中氨氮浓度与水温之间存在显著负相关;湘江熬洲、乔口、鹿角三个断面水体中氨氮浓度与溶解氧与之间存在显著正相关。对湘江水环境因子与水体中氨氮、硝态氮、总无机氮的年际变异特征进行分析,研究结果表明:1990~2005年,湘江水体中氨氮、硝态氮、总无机氮浓度年际变化范围分别为0.01-1.50mg/L、0.02-1.28mg/L、0.26-2.29mg/L,硝态氮浓度随年际变化略有上升,大部分年份硝态氮年均浓度高于氨氮;氨氮、硝态氮、总无机浓度与水环境因子年际变化之间相关性均不显著。以湘江干流归阳至衡山段集雨控制区为研究范围,采用灰色系统理论预测法、指数平滑预测法、模糊线性回归预测法、神经网络预测法等四种能捕捉非线性变化规律的预测方法分别构造了控制区输出断面水体中总氮浓度及其有关影响因素的预测模型。以该断面水体中2002~2005年总氮浓度的预测值和已有的实测值为基础数据,以最大拟合误差值、平均误差、平均绝对误差、平均相对误差、平均相对误差绝对值、均方根误差、Theil不等系数等7个衡量预测方法精度的评价指标为依据,运用因子分析法对所建模型进行综合评判,优选出BP神经网络预测模型为该断面水体中总氮浓度未来变化预测的最优拟合模型。并对所优选出的模型进行预测效果分析,可知BP神经网络预测模型在水体总氮浓度预测中完全能满足实际应用对误差的要求,预测合格率为100%。研究表明湘江水体中来源于研究区域内人类活动所产生的非点源总氮负荷量与区域内单位耕地面积氮肥施用量相关性极显著(R=0.899);并且,研究发现水稻生产中总氮的径流损失总量(y,kg·hm-2)与氮施用量(x,kg·hm-2)存在极显著的线性相关:y=0.0087x+3.5248(r=0.9585,p=0.0036)。基于农业生产中氮肥施用量及施肥方法是影响氮流失的两个能被种植者控制的因子,本文开展水稻生产有机无机肥配合施用以及适宜生态、经济施氮量研究。有机无机肥配合施用大田试验研究结果表明:有机-无机肥料配合施用比纯化肥处理增产135.00-562.OOkg/hm2,增幅9.87%~23.68%;有机无机肥料配合施用比例对水稻产量影响较大,五个有机-无机肥料配合施用处理中,以施50%有机肥:50%无机肥的处理产量最高、达7050.00 kg/hm2,以施60%有机肥:40%无机肥用的处理平均氮素累积量、平均氮素回收率、平均氮素农学利用率最高,分别为:124.25 kg/hm2、31.44%、20.91%。适宜生态、经济施氮量大田试验结果表明:湘江流域水稻生产氮素经济最佳施用量为132.31kg/hm2;在估算出湘江干流归阳至衡山段集雨控制区域内水稻生产中氮肥施用的环境成本的基础上,得到水稻生产氮素生态效益最佳施肥量为129.31kg/hm2。

【Abstract】 Xiang Jiang is the mother river of Hunan province, and increasing, serious pollution of Xiang Jiang river has been paid much attention by the society. The nitrogen pollution of water has been mainly caused by agricultural non-point source pollution. A general analysis of the temporal and spatial variation characteristics of nitrogen in Xiang Jiang river and Dong Ting lake, and study on the amounts of rational nitrogen application of rice, can provide scientific foundation for avoiding eutrophication of water in Xiang Jiang river, inhibit nitrogen pollution in water, and select a better prediction technique of nitrogen concentration in Xiang Jiang river, and can also offer a new method for the pollution control and management. The main research contents and results from our studies are as follows.The annual variation characteristics of water environmental indexes (water temperature, pH value, dissolved oxygen) and total nitrogen, ammonia nitrogen, nitrate nitrogen of water in Xiang Jiang river and Dong Ting lake were analyzed and the results showed that, the water temperature and dissolved oxygen value in Xiang Jiang river and Dong Ting lake had obviously seasonal characteristics, and the change of pH value was not significant from April of 2006 to March of 2007. The concentrations of ammonia nitrogen, nitrate nitrogen and total nitrogen of water in Xiang Jiang river were 0.03~1.76 mg/L,0.02~1.68 mg/L and 1.11~7.62 mg/L respectively, and the concentrations of ammonia nitrogen, nitrate nitrogen and total nitrogen of water in Dong Ting lake were 0.03~1.56 mg/L,0.04~1.72 mg/L and 0.04-9.81 mg/L respectively. The effect of agricultural non-point source pollution on total nitrogen concentrations of the Xiang Jiang river and Dong Ting lake was higher with the peak values appearing in July and August. There was significant correlation between total nitrogen concentrations and water temperatures of AoZhou, QiaoKou sections in Xiang Jiang river and WanZi lake, MuPing lake in Dong Ting lake. There existed significantly, negative correlation between ammonia nitrogen concentrations and water temperatures of AoZhou, QiaoKou and LuJiao sections in Xiang Jiang river and WanZi lake, MuPing lake in Dong Ting lake. There was significant correlation between NH4+-N concentrations and dissolved oxygen values of AoZhou, QiaoKou and LuJiao sections in Xiang Jiang river.The annual variation characteristics of water environmental indexes and ammonia nitrogen, nitrate nitrogen, total inorganic nitrogen concentrations of water in Xiang Jiang river were analyzed and the results showed that, the changes of ammonia nitrogen, nitrate nitrogen and total inorganic nitrogen concentrations of water in Xiang Jiang river were 0.01~1.50 mg/L,0.02~1.28 mg/L,0.26~2.29 mg/L respectively during years of 1990~2005 and the nitrate nitrogen concentration was observed to be increased a little annually. In most of the years monitored, the average annual concentration of nitrate nitrogen was higher than that of ammonia nitrogen. There was no significant correlation between total inorganic nitrogen, nitrate nitrogen concentrations and water environmental indexes.The catchment area of Xiang Jiang mainstream from Guiyang to Hengshan was the main focus of our studies, and we adopted the grey system forecast model, triple exponential smoothing forecast model, fuzzy linear regression forecast model and BP neural network forecast model, which could efficiently describe nonlinear regularities of water body, to identify the relationship between the total nitrogen concentration and related factors. Based on the predicted and monitored values of total nitrogen concentrations, the prediction precision was assessed with seven indexes including root mean squared error, average error, average absolute error and so on. The BP neural network was found to be the best model to predict the water total nitrogen concentration and all the relative errors were less than 20% of the difference between predicted and measured values and the qualified rate of prediction was 100%.Our studies on the correlation between the fertilizer input per unit arable area and the non-point source TN load of anthropogenic activity contribution to Xiangjiang river showed that they had obvious linear relationship and the coefficient of correlation was 0.899. moreover, the studies on the effects of different fertilization treatments on total nitrogen loss with runoff,it showed that There was very significant correlation between them and the linear equation is y=0.0087χ+3.5248 (r=0.9585,p=0.0036).Based on controlled nitrogen fertilizer application amount and application method to increase the nitrogen use efficiency in agriculture production, the combined application of organic and inorganic fertilizers and ecological and economical nitrogen application for rice were studied. The results from combined application of organic and inorganic fertilizers in the rice field experiment demonstrated that, the yield from combined application was 135.00-562.00 kg/hm2 higher than that from pure nitrogen application treatment, with an increase of 9.87~23.68%. The effect of combined application of organic and inorganic fertilizer on yield of rice was bigger, and the highest yield of rice among the five combined application treatments was from the 50% organic fertilizer and 50% inorganic fertilizer treatment with a yield of 7050.00 kg/hm2. The average nitrogen accumulation amount, nitrogen recovery percent and nitrogen agriculture rate of the 60% organic fertilizer and 40% inorganic fertilizer treatment were the highest, and they were 124.25 kg/hm2,31.44%,20.91% respectively. The results of ecological and economical nitrogen application experiment in the field showed that, the economical nitrogen application of rice cultivation in Xiang Jiang valley was 132.31 kg/hm2. Based on the estimate of the environmental cost of nitrogen application in rice cultivation between GuiYang and HengShan sections of Xiang Jiang river, it could be concluded that the best nitrogen application amount for the ecological and economical nitrogen benefit is 129.31 kg/hm2.

【关键词】 湘江氮素预测方法水稻施氮
【Key words】 Xiangjiang rivernitrogenprediction techniquericenitrogen application
节点文献中: 

本文链接的文献网络图示:

本文的引文网络