Determination of nitrogen and water stress with hyper spectral reflectance on maize using classification tree (CT) analysis

Abstract

Field experiments were conducted at the Tamil Nadu Agricultural University, Coimbatore, Tamilnadu, India during rabi (winter) season of 2013-14 with maize crop (TNAU maize hybrid Co 6). To ensure the stressed environment, the crop was subjected to two irrigation levels (IW/CPE: 0.80 and 0.50) and five staggered nitrogen levels (0, 50, 75, 100 and 125 % of recommended dose of nitrogen (RDN).The experiment was laid out in factorial randomized blocks design RBD (Factorial) with three replications. Hyper spectral observations were made with spectroradiometer GER 1500 at 60 and 90 days after sowing (DAS). Measured spectral reflectance curve of maize exhibited a broad low intensity peak centered in the green region at 550 nm and a sharp rise starting at about 675 nm to a plateau in the vicinity of 762 nm under unstressed environment created with irrigation at 0.80 IW/CPE ratio and fertilizer application at 100 % RDN. Significant differences in reflectance were established for nitrogen and water stress at green and NIR region. The results of Classification tree analysis revealed that nitrogen and water stress can be assessed and differentiated using reflectance data when transformed into spectral vegetation indices viz., NDVI, GNDVI, RVI, LCI, IR-RED and SR. Further the Classification tree algorithm coulddetermine that NDVI was the most effective index to assess the combined effect of nitrogen and water stress in maize crop.

Using hyperspectral leaf reflectance to estimate photosynthetic capacity and nitrogen content across eastern cottonwood and hybrid poplar taxa

“…Ye et al [ 47 ] found that when leaf N content increased, spectral reflectance decreased, and this relationship was more significantly observed in the green to near infrared regions. Due to these relationships, if robust statistical models can be developed, hyperspectral leaf reflectance can be used as a proxy for rapid assessment of photosynthetic capacity [ 48 – 50 ], as well as non-destructive and real-time monitoring of leaf N content [ 51 , 52 ]. On the other hand, creating empirical models for hyperspectral data requires sophisticated model development techniques, and model robustness depends on the quality of sample data measured in the field [ 53 ].…”