天然草地牧草营养品质的高光谱遥感研究进展  

Hyperspectral remote sensing progress for forage nutritional quality and quantity in natural grassland

作　　者：高金龙[1] 刘洁 殷建鹏 葛静[1] 侯蒙京 冯琦胜[1] 梁天刚[1] GAO Jin-long;LIU Jie;YIN Jian-peng;GE Jing;HOU Meng-jing;FENG Qi-sheng;LIANG Tian-gang(State Key Laboratory of Grassland Agro-Ecosystem, Key Laboratory of Grassland Livestock Industry Innovation, Ministry of Agriculture, College of Pastoral Agriculture Science and Technology, Lanzhou University, Lanzhou 730020, China)

机构地区：[1]兰州大学草地农业生态系统国家重点实验室兰州大学农业农村部草牧业创新重点实验室兰州大学草地农业科技学院

出　　处：《草业学报》2020年第2期172-185,共14页Acta Prataculturae Sinica

基　　金：国家自然科学基金(31672484、31702175、41401472);长江学者和创新团队发展计划(IRT 17R50);国家“十三五”重点研发计划项目(2017YFC0504801);111计划(B12002)资助

摘　　要：天然草地牧草营养品质的优劣不仅影响家畜的生长发育,同时也影响畜产品的品质,对草牧业的发展具有至关重要的意义。高光谱遥感技术的飞速发展使深入研究天然草地牧草品质的动态变化成为可能。本研究综述了目前可利用的高光谱遥感数据以及天然草地牧草营养品质遥感反演的主要成果、常用方法和最新研究动态,分析了我国在天然草地牧草营养品质监测与评价方面尚存在数据获取困难、相关研究缺乏、软硬件性能不足等问题;在多种观测平台及相关技术不断革新背景下,探索星载、机载和地面高光谱数据的有机结合,强化高光谱遥感仪器性能,提高关键营养成分的反演精度是未来研究的重点。Evaluation of the nutritional quality and quantity of natural grassland forage is central in the optimization of grassland animal husbandry systems,and provides important information to enhance the quality of animal products and growth and development of livestock.Therefore,rapid and accurate estimates of forage nutrient content will greatly assist the rational utilization and effective management of natural alpine grassland.The rapid development of hyperspectral remote sensing technology shows promise of providing a tool for the dynamic monitoring and evaluation of grassland forage nutrition quality and quantity.This paper reviews the available hyperspectral remote sensing systems,such as Hyperion,the compact high resolution imaging spectrometer(CHRIS),RapidEye,WorldView-2,Sentinel-2,HJ-1A hyper-spectrum imager(HSI),Tiangong-1,Gaofen-5 and Gaofen-6.We suggest that research progress in grassland nutrition will be greatly improved by making full use of these available data gathering resources.The commonly used remote sensing inversion methods for forage nutrient composition of natural grassland include traditional multivariate statistical analysis,regression analysis based on spectral features and spectral indexes,and physical modeling.Previous studies of the estimation and evaluation of key nutrients in forage focused on analysis of hyperspectral characteristics,spatial-temporal inversion models,spectral band selection and local area nutrient mapping.It is emphasized that the following issues remain:difficulties accessing data,lack of relevant research and performance deficiencies of existing software and hardware.Given the multiple observation platforms and related technology innovations,an important research direction is to improve the performance of hyperspectral instruments,and to improve the inversion accuracy of algorithms for determining key nutrients in future.Moreover,aerial and satellite data can be integrated to monitor the temporal and spatial nutritional changes at a landscape level in in natural gra

关 键 词：天然草地 牧草营养 品质 高光谱遥感 进展 

分 类 号：F32[经济管理—产业经济]