遥感技术已经越来越多地用于监测大面积植株的氮素状况和精准氮素养分管理。氮营养指数（nitrogen nutrition index，NNI）可以定量描述作物的氮素状况。然而，基于无人机多光谱影像的棉花NNI诊断尚缺乏研究。本研究评估了支持向量机（support vector machine，SVM）、反向传播神经网络（back propagation neural network，BPNN）和极端梯度提升（extreme gradient boosting，XGB）三种机器学习模型基于无人机多光谱影像估测棉花全生育期叶片氮含量和NNI的性能。研究结果表明，与氮含量和NNI相关性最高的前15个植被指数作为输入时，模型表现更优。三种模型中XGB模型在估测氮含量方面表现最优。上半冠层水平下的氮含量估测精度（率定集R²=0.89，RMSE=0.68 g m^-2，RE=14.62%；验证集R²=0.83，RMSE=1.08 g m^-2，RE=19.71%）远高于全冠层水平（率定集R²=0.73，RMSE=2.20 g m^-2，RE=26.70%；验证集R²=0.70，RMSE=2.48 g m^-2，RE=31.49%）和植株水平（率定集R²=0.66，RMSE=4.46 g m^-2，RE=30.96%；验证集R²=0.63，RMSE=3.69 g m^-2，RE=24.81%）。与之类似， XGB模型（率定集R²=0.65，RMSE=0.09，RE=8.59%；验证集R²=0.63，RMSE=0.09，RE=8.87%）在估测NNI方面也优于SVM 模型（率定集R²=0.62，RMSE=0.10，RE=7.92%； 验证集R²=0.60，RMSE=0.09，RE=8.03%）和BPNN模型（率定集R²=0.64，RMSE=0.09，RE=9.24%；验证集R²=0.62，RMSE=0.09，RE=8.38%）。基于最优XGB模型生成的NNI预测图，可以直观地诊断棉田氮素营养的空间分布和动态过程。本研究可以帮助农户及时、准确地实施棉花氮素精准管理。

Variants in chicken growth hormone receptor (GHR) gene lead to sex-linked dwarf (SLD) chickens, but effects of different variants are distinct. In this study, 11 SLD chicken breeds or strains including 3 Chinese native breeds and 8 breeding strains were studied in order to investigate the effects of different sex-linked dwarf variations on growth performance. The results showed that there were three reasons which could lead to dwarfism in the 11 breeds or strains. Firstly, an about 1.7 kb deletion of growth hormone receptor (GHR) gene leads to dwarfism in Jiangxi dwarf chicken, strains GF24, GF26, N308, N309, and N310. Secondly, a T354C mutation in exon 5 of the GHR gene leads to dwarfism in strains N301 and N305. Thirdly, an unknown variant leads to dwarfism in Guizhou Yellow Dwarf chicken and Yixing Bantam chicken. In addition, all individuals of N303 had the 1.7 kb deletion of the GHR gene, and additionally, some of them also carried the T354C mutation. As far as the performance of individuals were compared among T354C homozygote, deletion homozygote, and heterozygote carrying both T354C and deletion, it was found that the T354C’s impacts on body weight of Chinese chickens were maximum, the body weight of chickens with homozygote T354C was 92.12% of those with heterozygote, and the difference of the body weight between deletion homozygote and heterozygote was not significant. There was no significant difference of shank length among three genotypes.