| 基于FY-3D/MERSI数据的东北地区干旱监测方法研究 |
| |
| 引用本文： | 王岩,王敬宜,冯锐,李嘉宁,武晋雯,许常华,林毅,纪瑞鹏,于文颖,汪利诚.基于FY-3D/MERSI数据的东北地区干旱监测方法研究[J].干旱地区农业研究,2023(4):289-297. |
| |
| 作者姓名： | 王岩 王敬宜 冯锐 李嘉宁 武晋雯 许常华 林毅 纪瑞鹏 于文颖 汪利诚 |
| |
| 作者单位： | 沈阳建筑大学交通与测绘工程学院，辽宁 沈阳 110168;沈阳建筑大学交通与测绘工程学院，辽宁 沈阳 110168；中国气象局沈阳大气环境研究所，辽宁 沈阳 110166;中国气象局沈阳大气环境研究所，辽宁 沈阳 110166；辽宁省农业气象灾害重点实验室，辽宁 沈阳 110166;辽宁省气象服务中心，辽宁 沈阳 110166;锦州市气象局，辽宁 锦州 121000 |
| |
| 基金项目： | 风云卫星应用先行计划（FY-APP-2021.0302）；辽宁省教育厅研究项目（lnjc202015）；辽宁省民生科技计划项目（2021JH2/10200024）；沈阳市中青年科技创新人才项目（RC210431） |
| |
| 摘 要： | 干旱是影响东北地区粮食安全的主要农业气象灾害之一，遥感技术是一种可便捷进行大范围干旱监测的手段。针对目前遥感干旱指数在作物生长发育过程中监测干旱的局限性和适用性等问题，以东北地区玉米和大豆等主要大田作物发育期为切入点，基于FY-3D/MERSI卫星遥感数据和地面土壤相对湿度实测数据，开展不同作物发育阶段干旱监测指数适用性分析，结合径向基神经网络方法，构建全时期和分时期土壤相对湿度反演模型，利用实测土壤相对湿度数据开展精度验证与对比分析。结果表明：风云三号MERSI传感器数据在干旱监测中具有可行性，表观热惯量(ATI)在低植被覆盖或裸土时效果较好，适用于作物冻土期、裸土期和播种～拔节期；水分指数(WI)适用于播种～拔节期、拔节～抽雄期和成熟期等植被生长时期；分时期土壤相对湿度反演模型精度高于全时期土壤相对湿度反演模型，前者监测精度在80.0%以上，比全时期模型精度提高了10%～25%,尤其在冻土期(3月),分时期模型反演精度达到了92.6%。基于作物生长时期和形态差异，选择最适宜遥感干旱指数建立分时期土壤相对湿度反演模型，提高了干旱监测的准确性和可靠性。
|
| 关 键 词： | 风云卫星 遥感指数 干旱监测 径向基函数神经网络 模型适用性 东北地区 |
Drought monitoring method in Northeast China based on FY-3D/MERSI data |
| |
| WANG Yan,WANG Jingyi,FENG Rui,LI Jianing,WU Jinwen,XU Changhu,LIN Yi,JI Ruipeng,YU Wenying,WANG Licheng.Drought monitoring method in Northeast China based on FY-3D/MERSI data[J].Agricultural Research in the Arid Areas,2023(4):289-297. |
| |
| Authors: | WANG Yan WANG Jingyi FENG Rui LI Jianing WU Jinwen XU Changhu LIN Yi JI Ruipeng YU Wenying WANG Licheng |
| |
| Affiliation: | School of Transportation and Geomatics Engineering, Shenyang Jianzhu University, Shenyang, Liaoning 110168, China;School of Transportation and Geomatics Engineering, Shenyang Jianzhu University, Shenyang, Liaoning 110168, China; Institute of Atmospheric Environment, China Meteorological Administration, Shenyang, Liaoning 110166, China;Institute of Atmospheric Environment, China Meteorological Administration, Shenyang, Liaoning 110166, China; Key Laboratory of Agrometeorological Disasters, Liaoning Province, Shenyang, Liaoning 110166, China;Liaoning Provincial Meteorological Service Center, Shenyang, Liaoning 110166, China;Jinzhou Meteorological, Jinzhou, Liaoning 121000, China |
| |
| Abstract: | Drought is one of the main agrometeorological disasters affecting food security in Northeast China. Remote sensing technology provides a convenient and effective means for large\|scale drought monitoring. In view of the limitations and applicability of remote sensing drought index in drought monitoring during crop growth and development, taking the crop development period of main field crops in Northeast China as the starting point, such as corn and soybean, based on FY-3D/MERSI satellite remote sensing data and ground relative soil moisture measured data, the applicability analysis of drought monitoring index in different crop development stages was carried out. Combined with radial basis function neural network method, the inversion models of relative soil moisture in the whole period and different periods were constructed, and the accuracy verification and comparative analysis were carried out by using the measured relative soil moisture data. The results showed that FY-3D/MERSI data were feasible in drought monitoring, and the apparent thermal inertia was better in low vegetation coverage or bare soil, which was suitable for frozen soil period, bare soil period and sowing~jointing period. The water index was suitable for vegetation growth periods such as sowing~jointing, jointing~tasseling and maturity. The accuracy of the relative soil moisture inversion model in different periods was better than that of the relative soil moisture inversion model in the whole period. The monitoring accuracy was above 80.0%, which was 10%~25% higher than that of the whole period model. Especially in the frozen soil period (March), the inversion accuracy reached 92.6%. According to the difference of crop growth period and morphology, the suitable remote sensing drought index was selected, and the inversion model of relative soil moisture in different periods was established to improve the accuracy and reliability of drought monitoring. |
| |
| Keywords: | FY meteorological satellite remote sensing index drought monitoring radial basis function neural network (RBFNN) models applicability Northeast China |
|
| 点击此处可从《干旱地区农业研究》浏览原始摘要信息 |
|
点击此处可从《干旱地区农业研究》下载全文 |
|
