Modeling Managed Grassland Biomass Estimation by Using Multitemporal Remote Sensing Data-A Machine Learning Approach
Modeling Managed Grassland Biomass Estimation by Using Multitemporal Remote Sensing Data-A Machine Learning Approach
复制标题
DOI：
10.1109/jstars.2016.2561618
复制
发表时间：
2017-07-01
影响因子：
5.5
通讯作者：
Green, Stuart
中科院分区：
文献类型：
--
作者：
Ali, Iftikhar;Cawkwell, Fiona;Green, Stuart
在爱尔兰，超过80%的农业用地是草地，这是基于牧场的奶牛养殖和畜牧业的主要饲料来源。在全球范围内已经进行了许多研究，利用卫星遥感数据来估计草地生物量，但很少在像爱尔兰这样的集中管理的小规模牧场系统中进行，在这些牧场上，草既被放牧，也被收获作为冬季饲料。应用多元线性回归（MLR）、人工神经网络（ANN）和自适应神经模糊推理系统（ANFIS）模型对爱尔兰两个集约化管理草地农场的草地生物量（kg干物质/ha/d）进行了估算。第一个试验场（Moorepark）12年（2001-2012年）和第二个试验场（Grange）6年（2001-2005年，2007年）的原位测量（每周测量的生物量）用于模型开发。五个植被指数加上两个原始光谱波段（红色=红色波段，NIR=近红外波段）来自8天的MODIS产品（MOD 09 Q1）被用作所有三个模型的输入。模型评估表明，ANFIS（R-Moorepark（2）= 0.85，RMSEMoorepark = 11.07; R-Grange（2）= 0.76，RMSEGrange = 15.35）产生了改进的生物量估计相比，人工神经网络和MLR。所提议的方法将有助于更好地探索未来来自空间传感器的遥感数据的流入，以用于检索不同的生物物理参数，随着卫星系列新成员（ALOS-2、Radarsat 2、Sentinel、TerraSAR-X、TanDEM-X/L）的发射，开发处理大量图像数据的工具将变得日益重要。
More than 80% of agricultural land in Ireland is grassland, which is a major feed source for the pasture based dairy farming and livestock industry. Many studies have been undertaken globally to estimate grassland biomass by using satellite remote sensing data, but rarely in systems like Ireland's intensively managed, but small-scale pastures, where grass is grazed as well as harvested for winter fodder. Multiple linear regression (MLR), artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) models were developed to estimate the grassland biomass (kg dry matter/ha/day) of two intensively managed grassland farms in Ireland. For the first test site (Moorepark) 12 years (2001-2012) and for second test site (Grange) 6 years (2001-2005, 2007) of in situ measurements (weekly measured biomass) were used for model development. Five vegetation indices plus two raw spectral bands (RED=red band, NIR=Near Infrared band) derived from an 8-day MODIS product (MOD09Q1) were used as an input for all three models. Model evaluation shows that the ANFIS (R-Moorepark(2) = 0.85, RMSEMoorepark = 11.07; R-Grange(2) = 0.76, RMSEGrange = 15.35) has produced improved estimation of biomass as compared to the ANN and MLR. The proposed methodology will help to better explore the future inflow of remote sensing data from spaceborne sensors for the retrieval of different biophysical parameters, and with the launch of new members of satellite families (ALOS-2, Radarsat2, Sentinel, TerraSAR-X, TanDEM-X/L) the development of tools to process large volumes of image data will become increasingly important.