基于NDVI时间序列数据的施肥方式遥感识别方法
|
摘要
农产品生产过程时空动态监测是有机/绿色农产品认证亟待解决的问题,不同施肥方式的时空精准识别是解决该问题的关键。本文以美国加州大学戴维斯分校长期定位实验为基本材料,利用时间序列Landsat8和Sentinel-2影像研究长期施肥实验下不同施肥处理轮作地块的植被指数时间序列,对比分析不同施肥处理NDVI的差异以及NDVI与产量的相关性。结果表明:1)不同施肥处理下的NDVI时间序列曲线总体趋势相似,有机肥与化肥处理NDVI时间序列曲线差异较大;2)不同施肥处理NDVI随作物生长期呈现规律变化,生长初期和后期有机肥处理NDVI均值高于化肥处理,生长中期化肥处理高于有机肥处理;3)不同施肥处理下的NDVI与产量之间相关系数随作物生长期有规律变化,应用植被指数进行遥感估产需要考虑不同施肥处理的影响。研究成果初步探讨了利用不同施肥处理NDVI时间序列差异、NDVI与产量相关性差异区分有机肥与其他施肥方式,有望为有机/绿色农业的时空动态监测与认证提供遥感技术支持,深化遥感技术在农业领域应用。
Due to the effects of organic fertilizer and chemical fertilizer release rate and different fertilizers on soil physical and chemical properties, there are differences in response of crops to water and fertilizer in the process of growth. Therefore, time series remote sensing monitoring is needed to realize dynamic monitoring of space-time in different fertilization treatments. Remote sensing technology, as a means to rapidly acquire spatial and temporal dynamic surface information in a wide range, has been widely recognized as important for the development of modern agriculture with high yield, high efficiency and environmental friendliness. However, at present, remote sensing technology can only assist in monitoring the quality of small-scale agricultural products, while in the production process of organic agricultural products. The research on accurate identification of large-scale fertilization methods is still lacking. Spatio-temporal dynamic monitoring of agricultural production process is an urgent problem to be solved in organic/green agricultural product certification. Spatio-temporal accurate identification of different fertilization methods is the key to solve this problem. California Central Valley has Mediterranean climate, hot summer and little rain. Its unique climate conditions provide a good climate condition for acquiring remote sensing images of the whole growth period. The experimental plots in this area are independent and large in area(0.4 hm2). This provides a reference for monitoring crop growth using remote sensing images. Taking the long-term positioning experiment of University of California at Davis as the basic material and maize and tomato as the research objects under the long-term positioning experiment of different fertilization treatments, this paper uses Landsat 8 and Soleno-2 image of time series to study the rotation of fertilizer, fertilizer + green manure, organic manure + green manure in three different treatments. The time series of vegetation index in the plot is used to compare and analyze the difference of NDVI among different fertilization treatments and the correlation between NDVI and yield. The results show that: 1) the general trends of NDVI time series curves under different fertilization treatments are similar, and the difference between organic fertilizer and chemical fertilizer treatment NDVI time series curve is obvious; 2) the NDVI of different fertilization treatments changes regularly with crop growth period. The mean value of NDVI in organic fertilizer treatment is higher than that in chemical fertilizer treatment at the initial and late growth stage, and lower than that in chemical fertilizer treatment in the middle growth stage; 3) the correlations coefficients between NDVI and yield change regularly with crop growth period under different fertilization treatments, and the effects of different fertilization treatments should be considered when applying vegetation index to estimating yield by remote sensing; 4) Fertilizer + green manure application methods can ensure agricultural sustainability while obtaining more. The difference of NDVI time series in different fertilization treatments has been proved in the research, in which we can also know the correlation difference of NDVI and yield. The results of research could provide remote sensing technology support for spatio-temporal dynamic monitoring and certification of organic/green agriculture, which could be used to distinguish organic fertilizer from other fertilization methods, and could deepen the application of remote sensing in agriculture.
Due to the effects of organic fertilizer and chemical fertilizer release rate and different fertilizers on soil physical and chemical properties, there are differences in response of crops to water and fertilizer in the process of growth. Therefore, time series remote sensing monitoring is needed to realize dynamic monitoring of space-time in different fertilization treatments. Remote sensing technology, as a means to rapidly acquire spatial and temporal dynamic surface information in a wide range, has been widely recognized as important for the development of modern agriculture with high yield, high efficiency and environmental friendliness. However, at present, remote sensing technology can only assist in monitoring the quality of small-scale agricultural products, while in the production process of organic agricultural products. The research on accurate identification of large-scale fertilization methods is still lacking. Spatio-temporal dynamic monitoring of agricultural production process is an urgent problem to be solved in organic/green agricultural product certification. Spatio-temporal accurate identification of different fertilization methods is the key to solve this problem. California Central Valley has Mediterranean climate, hot summer and little rain. Its unique climate conditions provide a good climate condition for acquiring remote sensing images of the whole growth period. The experimental plots in this area are independent and large in area(0.4 hm2). This provides a reference for monitoring crop growth using remote sensing images. Taking the long-term positioning experiment of University of California at Davis as the basic material and maize and tomato as the research objects under the long-term positioning experiment of different fertilization treatments, this paper uses Landsat 8 and Soleno-2 image of time series to study the rotation of fertilizer, fertilizer + green manure, organic manure + green manure in three different treatments. The time series of vegetation index in the plot is used to compare and analyze the difference of NDVI among different fertilization treatments and the correlation between NDVI and yield. The results show that: 1) the general trends of NDVI time series curves under different fertilization treatments are similar, and the difference between organic fertilizer and chemical fertilizer treatment NDVI time series curve is obvious; 2) the NDVI of different fertilization treatments changes regularly with crop growth period. The mean value of NDVI in organic fertilizer treatment is higher than that in chemical fertilizer treatment at the initial and late growth stage, and lower than that in chemical fertilizer treatment in the middle growth stage; 3) the correlations coefficients between NDVI and yield change regularly with crop growth period under different fertilization treatments, and the effects of different fertilization treatments should be considered when applying vegetation index to estimating yield by remote sensing; 4) Fertilizer + green manure application methods can ensure agricultural sustainability while obtaining more. The difference of NDVI time series in different fertilization treatments has been proved in the research, in which we can also know the correlation difference of NDVI and yield. The results of research could provide remote sensing technology support for spatio-temporal dynamic monitoring and certification of organic/green agriculture, which could be used to distinguish organic fertilizer from other fertilization methods, and could deepen the application of remote sensing in agriculture.
引文
[1]Xue L,Dai T,Cao W,et al.Monitoring Leaf Nitrogen Status in Rice with Canopy Spectral Reflectance[J].Agronomy Journal,2004,96(1):135-142.
[2]Gianquinto G,Orsini F,Fecondini M,et al.Amethodological approach for defining spectral indices for assessing tomatonitrogen status and yield[J].European Journal of Agronomy,2011,35(3):135-143.
[3]黄玉萍，Renfu Lu，戚超，陈坤杰.波长比和近红外光谱的番茄品质检测方法[J].光谱学与光谱分析，2018,38(8):2362-2368.Huang Yuping,Renfu Lu,Qi Chao,Chen Kunjie.Measurement of tomato quality attributes based on wavelength ratio and near-infrared spectroscopy[J].Spectroscopy and Spectral Analysis,2018,38(8):2362-2368.(in Chinese with English abstract)
[4]杨小玲.高光谱图像技术检测玉米种子品质研究[D].杭州：浙江大学，2016.Yang Xiaoling.Detection of Maize Seed Quality Using Hyperspectral Imaging[D].Hangzhou:Zhejiang University,2016.(in Chinese with English abstract)
[5]张新玉，王颖杰，刘若西，等.近红外光谱技术应用于玉米单籽粒蛋白质含量检测分析的初步研究[J].中国农业大学学报，2017,22(5):25-31.Zhang Xinyu,Wang Yingjie,Liu Ruoxi,et al.Application of near-infrared spectroscopy technology to analyze protein content in single kenel maize seed[J].Journal of China Agricultural University,2017,22(5):25-31.(in Chinese with English abstract)
[6]钟函笑，边金虎，李爱农.Landsat-8 OLI与Sentinel-2A MSI山区遥感影像辐射一致性研究[J].遥感技术与应用，2018,33(03):428-438.Zhong Hanxiao,Bian Jinhu,Li Ainong.Radiometric consistency between Landsat-8 OLI and Sentia-2 MSIimagery in mountainous terrain[J].Remote Sensing Technology and Application,2018,33(3):428-438.(in Chinese with English abstract)
[7]Feng Chen.A Preliminary investigation on comparison and transformation of Sentinel-2A MSI and Landsat-8 OLI[C].ISPRS Technical Commission III on Remote Sensing.Proceedings of the ISPRS Technical Commission III midterm symposium on"Developments,technologies and applications in remote sensing".ISPRS Technical Commission III on Remote Sensing,2018:6.
[8]郭昱杉，刘庆生，刘高焕，等.基于MODIS时序NDVI主要农作物种植信息提取研究[J].自然资源学报，2017,32(10):1808-1818.Guo Yushan,Liu Qingsheng,Liu Gaohuan,et al.Extraction of main crops in yellow river delta based on MODIS NDVItime series[J].Journal of Natural Resources,2017,32(10):1808-1818.(in Chinese with English abstract)
[9]王文静，张霞，赵银娣，等.综合多特征的Landsat-8时序遥感图像棉花分类方法[J].遥感学报，2017,21(1):115-124.Wang Wenjing,Zhang Xia,Zhao Yindi,et al.Cotton extraction method of integrated multi-features based on multitemporal Landsat-8 images[J].Journal of Remote Sensing,2017,21(1):115-124.(in Chinese with English abstract)
[10]韩衍欣，蒙继华，徐晋.基于NDVI与物候修正的大豆长势评价方法[J].农业工程学报，2017,33(2):177-182.Han Yanxin,Meng Jihua,Xu Jin.Soybean growth assessment method based on NDVI and phenological calibration[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2017,33(2):177-182.(in Chinese with English abstract)
[11]平跃鹏，臧淑英.基于MODIS时间序列及物候特征的农作物分类[J].自然资源学报，2016,31(3):503-513.Ping Yuepeng,Zang Shuying.Crop identification based on MODIS NDVI time-series data and phenological char acteristics[J].Journal of Natural Resources,2016,31(3):503-513.(in Chinese with English abstract)
[12]杨闫君，占玉林，田庆久，等.基于GF-1/WFVNDVI时间序列数据的作物分类[J].农业工程学报，2015,31(24):155-161.Yang Yanjun,Zhan Yulin,Tian Qingjiu,et al.Crop classification based on GF-1/WFV NDVI time series[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2015,31(24):155-161.(in Chinese with English abstract)
[13]Schell J A.Monitoring Vegetation Systems in the Great Plains with ERTS[J].Nasa Special Publication,1973,351:309.
[14]刘焕军，孟令华，张新乐，等.基于时间序列Landsat影像的棉花估产模型[J].农业工程学报，2015,31(17):215-220.Liu Huanjun,Meng Linghua,Zhang Xinle,et al.Estimation model of cotton yield with time series Landsat images[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2015,31(17):215-220.(in Chinese with English abstract)
[15]王利民，杨玲波，刘佳，等.GF-1和MODIS影像冬小麦长势监测指标NDVI的对比[J].作物学报，2018,44(7):1043-1054.Wang Limin,Yang Lingbo,Liu Jia,et al.Comparison of growth monitoring index NDVI between GF-1 and MODISimages in winter wheat[J].Acta Agronomica Sinica,2018,44(7):1043-1054.(in Chinese with English abstract)
[16]赵秉强，张夫道.我国的长期肥料定位实验研究[J].植物营养与肥料学报，2002,8(增刊)：3-8.Zhao Bingqiang,Zhang Fudao.Long-term fertilizer location test in China[J].Plant Nutrition and Fertilizer Science,2002,8(Supplement):3-8.(in Chinese with English abstract)
[17]沈善敏.长期土壤肥力实验的科学价值[J].植物营养与肥料学报，1995,1(1):1-9.Shen Shanmin.The scientific value of long-term fertility experiment[J].Plant Nutrition and Fertilizer Science,1995,1(1):1-9.
[18]Yang Z C,Zhao N,Huang F,et al.Long-term effects of different organic and inorganic fertilizer treatments on soil organic carbon sequestration and crop yields on the North China Plain[J].Soil&Tillage Research,2015,146:47-52.
[19]Yaduvanshi N P S.Effect of five years of rice-wheat cropping and NPK fertilizer use with and without organic and green manures on soil properties and crop yields in a reclaimed sodic soil[J].Journal of the Indian Society of Soil Science,2001,49:714-719.
[20]温延臣，李燕青，袁亮，等.长期不同施肥制度土壤肥力特征综合评价方法[J].农业工程学报，2015,31(7):91-99.Wen Tingchen,Li Yanqing,Yuan Liang,et al.Comprehensive assessment methodology of characteristics of soil fertility under different fertilization regimes in North China[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2015,31(7):91-99.(in Chinese with English abstract)
[21]焦险峰，杨邦杰，裴志远，等.基于植被指数的作物产量监测方法研究[J].农业工程学报，2005,21(4):104-108.Jiao Xianfeng,Yang Bangjie,Pei Zhiyuan,et al.Monitoring crop yield using NOAA/AVHRR-based vegetation indices[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2005,21(4):104-108.(in Chinese with English abstract)
[2]Gianquinto G,Orsini F,Fecondini M,et al.Amethodological approach for defining spectral indices for assessing tomatonitrogen status and yield[J].European Journal of Agronomy,2011,35(3):135-143.
[3]黄玉萍，Renfu Lu，戚超，陈坤杰.波长比和近红外光谱的番茄品质检测方法[J].光谱学与光谱分析，2018,38(8):2362-2368.Huang Yuping,Renfu Lu,Qi Chao,Chen Kunjie.Measurement of tomato quality attributes based on wavelength ratio and near-infrared spectroscopy[J].Spectroscopy and Spectral Analysis,2018,38(8):2362-2368.(in Chinese with English abstract)
[4]杨小玲.高光谱图像技术检测玉米种子品质研究[D].杭州：浙江大学，2016.Yang Xiaoling.Detection of Maize Seed Quality Using Hyperspectral Imaging[D].Hangzhou:Zhejiang University,2016.(in Chinese with English abstract)
[5]张新玉，王颖杰，刘若西，等.近红外光谱技术应用于玉米单籽粒蛋白质含量检测分析的初步研究[J].中国农业大学学报，2017,22(5):25-31.Zhang Xinyu,Wang Yingjie,Liu Ruoxi,et al.Application of near-infrared spectroscopy technology to analyze protein content in single kenel maize seed[J].Journal of China Agricultural University,2017,22(5):25-31.(in Chinese with English abstract)
[6]钟函笑，边金虎，李爱农.Landsat-8 OLI与Sentinel-2A MSI山区遥感影像辐射一致性研究[J].遥感技术与应用，2018,33(03):428-438.Zhong Hanxiao,Bian Jinhu,Li Ainong.Radiometric consistency between Landsat-8 OLI and Sentia-2 MSIimagery in mountainous terrain[J].Remote Sensing Technology and Application,2018,33(3):428-438.(in Chinese with English abstract)
[7]Feng Chen.A Preliminary investigation on comparison and transformation of Sentinel-2A MSI and Landsat-8 OLI[C].ISPRS Technical Commission III on Remote Sensing.Proceedings of the ISPRS Technical Commission III midterm symposium on"Developments,technologies and applications in remote sensing".ISPRS Technical Commission III on Remote Sensing,2018:6.
[8]郭昱杉，刘庆生，刘高焕，等.基于MODIS时序NDVI主要农作物种植信息提取研究[J].自然资源学报，2017,32(10):1808-1818.Guo Yushan,Liu Qingsheng,Liu Gaohuan,et al.Extraction of main crops in yellow river delta based on MODIS NDVItime series[J].Journal of Natural Resources,2017,32(10):1808-1818.(in Chinese with English abstract)
[9]王文静，张霞，赵银娣，等.综合多特征的Landsat-8时序遥感图像棉花分类方法[J].遥感学报，2017,21(1):115-124.Wang Wenjing,Zhang Xia,Zhao Yindi,et al.Cotton extraction method of integrated multi-features based on multitemporal Landsat-8 images[J].Journal of Remote Sensing,2017,21(1):115-124.(in Chinese with English abstract)
[10]韩衍欣，蒙继华，徐晋.基于NDVI与物候修正的大豆长势评价方法[J].农业工程学报，2017,33(2):177-182.Han Yanxin,Meng Jihua,Xu Jin.Soybean growth assessment method based on NDVI and phenological calibration[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2017,33(2):177-182.(in Chinese with English abstract)
[11]平跃鹏，臧淑英.基于MODIS时间序列及物候特征的农作物分类[J].自然资源学报，2016,31(3):503-513.Ping Yuepeng,Zang Shuying.Crop identification based on MODIS NDVI time-series data and phenological char acteristics[J].Journal of Natural Resources,2016,31(3):503-513.(in Chinese with English abstract)
[12]杨闫君，占玉林，田庆久，等.基于GF-1/WFVNDVI时间序列数据的作物分类[J].农业工程学报，2015,31(24):155-161.Yang Yanjun,Zhan Yulin,Tian Qingjiu,et al.Crop classification based on GF-1/WFV NDVI time series[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2015,31(24):155-161.(in Chinese with English abstract)
[13]Schell J A.Monitoring Vegetation Systems in the Great Plains with ERTS[J].Nasa Special Publication,1973,351:309.
[14]刘焕军，孟令华，张新乐，等.基于时间序列Landsat影像的棉花估产模型[J].农业工程学报，2015,31(17):215-220.Liu Huanjun,Meng Linghua,Zhang Xinle,et al.Estimation model of cotton yield with time series Landsat images[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2015,31(17):215-220.(in Chinese with English abstract)
[15]王利民，杨玲波，刘佳，等.GF-1和MODIS影像冬小麦长势监测指标NDVI的对比[J].作物学报，2018,44(7):1043-1054.Wang Limin,Yang Lingbo,Liu Jia,et al.Comparison of growth monitoring index NDVI between GF-1 and MODISimages in winter wheat[J].Acta Agronomica Sinica,2018,44(7):1043-1054.(in Chinese with English abstract)
[16]赵秉强，张夫道.我国的长期肥料定位实验研究[J].植物营养与肥料学报，2002,8(增刊)：3-8.Zhao Bingqiang,Zhang Fudao.Long-term fertilizer location test in China[J].Plant Nutrition and Fertilizer Science,2002,8(Supplement):3-8.(in Chinese with English abstract)
[17]沈善敏.长期土壤肥力实验的科学价值[J].植物营养与肥料学报，1995,1(1):1-9.Shen Shanmin.The scientific value of long-term fertility experiment[J].Plant Nutrition and Fertilizer Science,1995,1(1):1-9.
[18]Yang Z C,Zhao N,Huang F,et al.Long-term effects of different organic and inorganic fertilizer treatments on soil organic carbon sequestration and crop yields on the North China Plain[J].Soil&Tillage Research,2015,146:47-52.
[19]Yaduvanshi N P S.Effect of five years of rice-wheat cropping and NPK fertilizer use with and without organic and green manures on soil properties and crop yields in a reclaimed sodic soil[J].Journal of the Indian Society of Soil Science,2001,49:714-719.
[20]温延臣，李燕青，袁亮，等.长期不同施肥制度土壤肥力特征综合评价方法[J].农业工程学报，2015,31(7):91-99.Wen Tingchen,Li Yanqing,Yuan Liang,et al.Comprehensive assessment methodology of characteristics of soil fertility under different fertilization regimes in North China[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2015,31(7):91-99.(in Chinese with English abstract)
[21]焦险峰，杨邦杰，裴志远，等.基于植被指数的作物产量监测方法研究[J].农业工程学报，2005,21(4):104-108.Jiao Xianfeng,Yang Bangjie,Pei Zhiyuan,et al.Monitoring crop yield using NOAA/AVHRR-based vegetation indices[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2005,21(4):104-108.(in Chinese with English abstract)