基于无人机遥感技术的黄华占水稻施肥决策模型研究
|
摘要
目前,水稻生产中氮肥施用过量,肥料利用率和产量相对较低等问题日益突出。无人机遥感能够实现无损、及时、快速大面积地获取作物田间信息,已被广泛应用于精准农业管理中。以黄华占水稻为研究对象,设计了不同施氮梯度的小区试验,利用无人机搭载rededge-M多光谱相机获取水稻生育期冠层多光谱图像,提取NDVI植被指数,分析了不同氮素条件下水稻冠层NDVI值的变化规律,研究了NDVI和标准种植比值指数(RISP)与水稻植株氮含量之间的相关关系;依据有效积温数据,建立了基于标准种植比值法的水稻关键施肥节点的施肥量决策模型。研究结果表明:随施肥量增加,水稻冠层NDVI值也随之变大,整个生育期呈现出"快速增加-缓慢增加-缓慢降低"趋势;NDVI和RISP值均与水稻植株氮含量显著相关,且RISP的相关系数达0.9以上,可更好地用于诊断水稻生育期氮素养分状况;基于标准种植比值法的水稻施肥决策模型拟合决定系数为0.991。模型验证试验发现,模型种植区平均施氮用量为99.64kg·hm-2,而传统种植区平均用量为135.60kg·hm-2,施肥量减少26.52%。且模型种植和传统种植的产量差异不足1%,说明该施肥模型在保证产量的同时提高了氮肥利用率,为实现作物养分管理决策支持系统提供了一种新的模型方法。
Problems, such as excessive application of nitrogen fertilizer, relatively low fertilizer utilization rate and low yield in rice production are becoming increasingly prominent at present. Unmanned aerial vehicle(UAV) remote sensing can achieve cropless field information in a non-destructive, timely and fast manner, and has been widely used in precision agriculture management. In this paper, Huanghuazhan rice was used as the research object, and the experiment of different nitrogen application gradients was designed. The multiedge image of rice growth period was obtained by using the rededge-M multispectral camera. The NDVI vegetation index was extracted and the different nitrogen conditions were analyzed. The relationship between NDVI values of rice canopy and the ratio of NDVI to standard planting ratio index(RISP) and nitrogen content of rice plants were studied. Based on the effective accumulated temperature data and the key fertilization nodes of rice according to on the standard planting ratio method, The fertilization volume decision model was established. The results showed that he DNVI value of rice canopy increased with the increase of fertilization amount, and the whole growth period showed a trend of "rapid increase-slow increase-slow decrease". NDVI and RISP values were significantly correlated with nitrogen contents of rice plants.And the determination coefficient of RISP was above 0.9, which can be better used to diagnose nitrogen nutrient status during rice growth period. The fitting coefficient of rice fertilization decision model based on standard planting ratio method was 0.991. And the model verification experiment showed that the average amount of nitrogen used the model planting area was 99.64 kg·hm-2, while the average dosage in traditional planting areas was 135.60 kg·hm-2, with a reduction by 26.52%. The difference between the model planting and the traditional planting yield was less than 1%, indicating that the fertilization model improves the nitrogen fertilizer utilization rate while ensuring the yield, and provides a new model method for the crop nutrient management decision support system.
Problems, such as excessive application of nitrogen fertilizer, relatively low fertilizer utilization rate and low yield in rice production are becoming increasingly prominent at present. Unmanned aerial vehicle(UAV) remote sensing can achieve cropless field information in a non-destructive, timely and fast manner, and has been widely used in precision agriculture management. In this paper, Huanghuazhan rice was used as the research object, and the experiment of different nitrogen application gradients was designed. The multiedge image of rice growth period was obtained by using the rededge-M multispectral camera. The NDVI vegetation index was extracted and the different nitrogen conditions were analyzed. The relationship between NDVI values of rice canopy and the ratio of NDVI to standard planting ratio index(RISP) and nitrogen content of rice plants were studied. Based on the effective accumulated temperature data and the key fertilization nodes of rice according to on the standard planting ratio method, The fertilization volume decision model was established. The results showed that he DNVI value of rice canopy increased with the increase of fertilization amount, and the whole growth period showed a trend of "rapid increase-slow increase-slow decrease". NDVI and RISP values were significantly correlated with nitrogen contents of rice plants.And the determination coefficient of RISP was above 0.9, which can be better used to diagnose nitrogen nutrient status during rice growth period. The fitting coefficient of rice fertilization decision model based on standard planting ratio method was 0.991. And the model verification experiment showed that the average amount of nitrogen used the model planting area was 99.64 kg·hm-2, while the average dosage in traditional planting areas was 135.60 kg·hm-2, with a reduction by 26.52%. The difference between the model planting and the traditional planting yield was less than 1%, indicating that the fertilization model improves the nitrogen fertilizer utilization rate while ensuring the yield, and provides a new model method for the crop nutrient management decision support system.
引文
[1] GUTIERREZ RODRIGO.Systems biology for enhanced plant nitrogen nutrition[J].Science,2012,336(6089):1673-1675.
[2]蔡红光,米国华,张秀芝,等.不同施肥方式对东北黑土春玉米连作体系土壤氮素平衡的影响[J].植物营养与肥料学报,2012,18(1):89-97.
[3]陈鹏飞,孙九林,王纪华,等.基于遥感的作物氮素营养诊断技术:现状与趋势[J].中国科学(信息科学),2010(S1):21-37.
[4]刘建刚,赵春江,杨贵军,等.无人机遥感解析田间作物表型信息研究进展[J].农业工程学报,2016(24):98-106.
[5]赵春江.对我国未来精准农业发展的思考[J].农业网络信息,2010(4):5-8.
[6] COLOMINA I,MOLINA P.Unmanned aerial systems for photogrammetry and remote sensing:A review[J].Isprs Journal of Photogrammetry&Remote Sensing,2014,92(2):79-97.
[7]秦占飞,常庆瑞,谢宝妮,等.基于无人机高光谱影像的引黄灌区水稻叶片全氮含量估测[J].农业工程学报,2016,32(23):77-85.
[8]马明洋,许童羽,周云成,等.东北粳稻叶绿素相对含量的无人机高清影像检测方法[J].沈阳农业大学学报,2017,48(6):757-762.
[9] BALLESTEROS R,ORTEGA J F,HERNANDEZ D,et al.Applications of georeferenced high-resolution images obtained with unmanned aerial vehicles.Part II:application to maize and onion crops of a semi-arid region in Spain[J].Precision Agriculture,2014,15(6):593-614.
[10] SWAIN K C,THOMSON S J,JAYASURIYA H P W.Adoption of an unmanned helicopter for low-altitude remote sensing to estimate yield and total biomass of a rice crop[J].Transactions of the Asabe,2010,53(1):21-27.
[11]刘奕彤,宋玉柱,马昕宇,等.基于多光谱成像技术的玉米氮素营养诊断方法研究[J].农机化研究,2018,40(2):148-153.
[12]张雨.基于无人机遥感的水稻氮素营养诊断研究[D].哈尔滨:东北农业大学,2017.
[13]刘忠,万炜,黄晋宇,等.基于无人机遥感的农作物长势关键参数反演研究进展[J].农业工程学报,2018,34(24):60-71.
[14]姚霞,刘勇,王妮,等.基于无人机遥感的小麦氮素营养和生长监测[C]//2014年中国作物学会学术年会,2014.
[15]沈国权.影响作物发育速度的非线性温度模式[J].气象,1980(6):9-11.
[16]田茁.基于BP神经网络的测土配方施肥模型[J].黑龙江科技信息,2015(29):145-146.
[17]周炼清,史舟,王珂,等.基于WebGIS的农业园区水稻施肥推荐系统的研究[J].浙江大学学报,2005(2):60-64.
[18] GENG P. Study on fertilization model of nitrogen, phosphorus and potassium for high yield of rice[J]. Journal of Anhui Agricultural Sciences,2008,36(27):11852-11854.
[19] WRIGHT D L,RASMUSSEN V P,et al.Using remote sensing manage wheat grain protein[R].NASA SSC Report 2003,ARC-USU-001-02.
[20]杨玮,王秀,马伟,等.基于近地光谱探测技术的冬小麦变量施肥[J].吉林大学学报,2007,37(6):1455-1460.
[21]冯宗会.不同水氮管理下NDVI监测及优化施氮研究[D].北京:首都师范大学,2012.
[22]纪荣婷,闵炬,黄程鹏,等.光谱仪在作物施氮推荐中的应用研究进展———以GreenSeeker光谱仪为例[J].江苏农业科学,2017,45(2):9-13.
[23]殷丽萍,邹忠,王冬梅,等.水稻不同品种光谱监测试验研究[J].现代农业科技,2016(19):13-15.
[24]屠曾平,黄炽林,万霞.水稻发育预测及电脑的应用———介绍美国“DD50/25”发育预测法[J].广东农业科学,1985(4):9-12.
[2]蔡红光,米国华,张秀芝,等.不同施肥方式对东北黑土春玉米连作体系土壤氮素平衡的影响[J].植物营养与肥料学报,2012,18(1):89-97.
[3]陈鹏飞,孙九林,王纪华,等.基于遥感的作物氮素营养诊断技术:现状与趋势[J].中国科学(信息科学),2010(S1):21-37.
[4]刘建刚,赵春江,杨贵军,等.无人机遥感解析田间作物表型信息研究进展[J].农业工程学报,2016(24):98-106.
[5]赵春江.对我国未来精准农业发展的思考[J].农业网络信息,2010(4):5-8.
[6] COLOMINA I,MOLINA P.Unmanned aerial systems for photogrammetry and remote sensing:A review[J].Isprs Journal of Photogrammetry&Remote Sensing,2014,92(2):79-97.
[7]秦占飞,常庆瑞,谢宝妮,等.基于无人机高光谱影像的引黄灌区水稻叶片全氮含量估测[J].农业工程学报,2016,32(23):77-85.
[8]马明洋,许童羽,周云成,等.东北粳稻叶绿素相对含量的无人机高清影像检测方法[J].沈阳农业大学学报,2017,48(6):757-762.
[9] BALLESTEROS R,ORTEGA J F,HERNANDEZ D,et al.Applications of georeferenced high-resolution images obtained with unmanned aerial vehicles.Part II:application to maize and onion crops of a semi-arid region in Spain[J].Precision Agriculture,2014,15(6):593-614.
[10] SWAIN K C,THOMSON S J,JAYASURIYA H P W.Adoption of an unmanned helicopter for low-altitude remote sensing to estimate yield and total biomass of a rice crop[J].Transactions of the Asabe,2010,53(1):21-27.
[11]刘奕彤,宋玉柱,马昕宇,等.基于多光谱成像技术的玉米氮素营养诊断方法研究[J].农机化研究,2018,40(2):148-153.
[12]张雨.基于无人机遥感的水稻氮素营养诊断研究[D].哈尔滨:东北农业大学,2017.
[13]刘忠,万炜,黄晋宇,等.基于无人机遥感的农作物长势关键参数反演研究进展[J].农业工程学报,2018,34(24):60-71.
[14]姚霞,刘勇,王妮,等.基于无人机遥感的小麦氮素营养和生长监测[C]//2014年中国作物学会学术年会,2014.
[15]沈国权.影响作物发育速度的非线性温度模式[J].气象,1980(6):9-11.
[16]田茁.基于BP神经网络的测土配方施肥模型[J].黑龙江科技信息,2015(29):145-146.
[17]周炼清,史舟,王珂,等.基于WebGIS的农业园区水稻施肥推荐系统的研究[J].浙江大学学报,2005(2):60-64.
[18] GENG P. Study on fertilization model of nitrogen, phosphorus and potassium for high yield of rice[J]. Journal of Anhui Agricultural Sciences,2008,36(27):11852-11854.
[19] WRIGHT D L,RASMUSSEN V P,et al.Using remote sensing manage wheat grain protein[R].NASA SSC Report 2003,ARC-USU-001-02.
[20]杨玮,王秀,马伟,等.基于近地光谱探测技术的冬小麦变量施肥[J].吉林大学学报,2007,37(6):1455-1460.
[21]冯宗会.不同水氮管理下NDVI监测及优化施氮研究[D].北京:首都师范大学,2012.
[22]纪荣婷,闵炬,黄程鹏,等.光谱仪在作物施氮推荐中的应用研究进展———以GreenSeeker光谱仪为例[J].江苏农业科学,2017,45(2):9-13.
[23]殷丽萍,邹忠,王冬梅,等.水稻不同品种光谱监测试验研究[J].现代农业科技,2016(19):13-15.
[24]屠曾平,黄炽林,万霞.水稻发育预测及电脑的应用———介绍美国“DD50/25”发育预测法[J].广东农业科学,1985(4):9-12.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |