Abstract: Determining the optimal estimation model of maize leaf area index (LAI) and above-ground biomass based on UAV multispectral remote sensing plays a significant role in obtaining timely, non-destructive, and reliable growth parameters. This study set four nitrogen treatments with Zhengdan 958 (ZD958) and Xianyu 335 (XY335) as materials from 2018 to 2019.   Multispectral images were acquired by a drone equipped with a multispectral camera, and the relationship between LAI and aboveground biomass and vegetation index was analyzed, and a vegetation index-based LAI and aboveground biomass prediction model was constructed. The results showed that the response of the same vegetation index to the change of nitrogen application was different between two varieties. At silking stage, the power equation received the best ranking in estimating the LAI and above-ground biomass for ZD958, and the exponential and power equation received the best ranking in estimating LAI and above-ground biomass for XY335, respectively. At filling stage, the power equation received the best ranking of fitting goodness in estimating LAI, while exponential did in estimating above-ground biomass for two maize varieties. This study provided an important basis and technical method for further improving the accuracy of monitoring growth parameters for spring maize.

Key words: spring maize, leaf area index, above-ground biomass, multispectral remote sensing, vegetation index, regression model

摘要： 明确基于无人机多光谱遥感的玉米叶面积指数（LAI）和地上部生物量的最优估算模型对获取即时、无损、可靠的长势关键参量具有重要意义。2018—2019年，以郑单958（ZD958）和先玉335（XY335）为研究对象，设置4个施氮处理，通过无人机搭载多光谱相机获取多光谱影像，分析两品种LAI和地上部生物量与植被指数相关性，分别构建了基于植被指数的LAI和地上部生物量预测模型。结果表明：同一植被指数在两品种中对施氮量的变化响应规律不同；在吐丝期,幂函数对ZD958的LAI和地上部生物量估算效果最好，指数函数对XY335的LAI估算效果好，幂函数对地上部生物量估算效果好；在灌浆期，幂函数对两品种的LAI估算效果最佳，而指数函数对两品种的地上部生物量估算效果最好。研究结果为进一步提高春玉米长势监测的精度提供了重要依据。

关键词: 春玉米, 叶面积指数, 地上部生物量, 多光谱遥感, 植被指数, 回归模型