Remote estimation of nitrogen and chlorophyll contents in maize at leaf and canopy levels

Highlights

• Leaf nitrogen and chlorophyll content were retrieved accurately from leaf reflectance spectra.

• Canopy nitrogen and chlorophyll content were closely related.

• Vegetation indices using green and red-edge spectral bands were used for accurate chlorophyll and nitrogen estimation at canopy level.

• Optimal spectral bands found for nitrogen and chlorophyll estimation match well-spectral bands of near future space systems.

Abstract

Leaf and canopy nitrogen (N) status relates strongly to leaf and canopy chlorophyll (Chl) content. Remote sensing is a tool that has the potential to assess N content at leaf, plant, field, regional and global scales. In this study, remote sensing techniques were applied to estimate N and Chl contents of irrigated maize (Zea mays L.) fertilized at five N rates. Leaf N and Chl contents were determined using the red-edge chlorophyll index with R² of 0.74 and 0.94, respectively. Results showed that at the canopy level, Chl and N contents can be accurately retrieved using green and red-edge Chl indices using near infrared (780–800 nm) and either green (540–560 nm) or red-edge (730–750 nm) spectral bands. Spectral bands that were found optimal for Chl and N estimations coincide well with the red-edge band of the MSI sensor onboard the near future Sentinel-2 satellite. The coefficient of determination for the relationships between the red-edge chlorophyll index, simulated in Sentinel-2 bands, and Chl and N content was 0.90 and 0.87, respectively.

Introduction

Numerous leaf-level studies have demonstrated a strong link between nitrogen (N) content and photosynthetic activity (Field and Mooney, 1986, Wullschleger, 1993). Kergoat et al. (2008) analyzed the relationship between leaf N content and the eddy covariance CO₂ flux measurements obtained at a range of diverse sites located in mid to high latitudes, encompassing managed and unmanaged stands, mono- and pluri-specific canopies. They concluded that leaf N content was a strong factor influencing both optimum canopy light use efficiency and canopy photosynthesis rate. Gitelson et al., 2003a, Gitelson et al., 2006a found a close, consistent relationship between gross primary productivity (GPP) and total plant Chl content in maize (Zea mays L.) and soybean (Glycine max (L.) Merr.). Moreover, it was shown that despite great differences in leaf structure and canopy architecture in the C₃ and C₄ crops studied, the relationship between GPP and Chl content was not species specific. Thus, monitoring of N and Chl content provides important information about crop photosynthetic status.

Evans, 1983, Evans, 1989 demonstrated the partitioning of N between protein fractions of soluble and thylakoid proteins remained unaltered with increasing N content. Therefore, changes in leaf N content will result in similar changes to the thylakoid pigment protein complex, that consists primarily of Chl, and the carbon fixing soluble protein enzyme activity of ribulose 1,5-bisphosphate (RuBP). An accurate measure of one component can provide estimates of the other two. Walters (2003) found a very close relationship between leaf Chl and N contents in maize, and also between leaf N content and crop yields. Baret et al. (2007) found that canopy Chl content was well suited for quantifying canopy-level N content. They concluded that canopy Chl content was a physically sound quantity that represents the optical path in the canopy where absorption by Chl dominates the radiometric signal. Thus, absorption by Chl provides the necessary link between remote sensing observations and canopy-state variables that are used as indicators of N status and photosynthetic capacity.

Reflectance in the green and red-edge spectral regions was shown to be optimal for non-destructive estimation of leaf Chl content in a wide range of its variation (Blackburn, 2006, Gitelson et al., 2003b, Gitelson, 2011a, Hatfield et al., 2008, le Maire et al., 2004, Richardson et al., 2002, Ustin et al., 2009). Féret et al. (2011), using a large set of leaf data collected from all over the world, showed that the Prospect 5 radiative transfer model provided an accurate estimation of leaf Chl content using reflectance in the red-edge and near infrared spectral regions.

However, at canopy level the estimation of Chl and N content is much more problematic and only a few papers have demonstrated a way to estimate them (Gitelson et al., 2005, Gitelson, 2011b, Lee et al., 2008, Sripada et al., 2008, Takahashi et al., 2000, Tian et al., 2011, Xue et al., 2004, Zhu et al., 2008). Among recently published results, it is worthy to note technique for the accurate assessment of N content in rice using hyperspectral measurements (Inoue et al., 2012). They found that the combination of reflectance values at two wavebands (near infrared at 825 nm and red-edge around 735 nm) has a significant and consistent role in the assessment of rice N content. High predictive accuracy was achieved when using models involving the normalized difference or the ratio of reflectance at these wavebands. Clevers and Kooistra (2012), using PROSAIL simulations, showed that the red-edge chlorophyll index (CI_red-edge, Table 1) with a wide red-edge band (700–730 nm) was linearly related to the canopy Chl content over the full range of potential Chl values. At their study sites, the CI_red-edge was also found to be a good and linear estimator of canopy N content in both grassland and potato cropping systems.

Clevers and Gitelson (2013) focused on the potential of Sentinel-2 (http://www.esa.int/Our_Activities/Observing_the_Earth/GMES/Sentinel-2) and Sentinel-3 (http://www.esa.int/Our_Activities/Observing_the_Earth/GMES/Sentinel-3) satellites for estimating total crop and grassland Chl and N contents. The Multi Spectral Instrument (MSI) on the Sentinel-2 satellite system has bands centered at 560 nm (green), 665 nm (red), 705 and 740 nm (red edge) and 783 nm (NIR). The Ocean and Land Color Instrument (OLCI) on the Sentinel-3 satellite system has one spectral band in red edge region centered at 709 nm. Clevers and Gitelson (2013) found that the CI_red-edge, the green chlorophyll index, CI_green, and the MERIS terrestrial chlorophyll index, MTCI (Table 1) were accurate and linear estimators of canopy Chl and N contents. Bands of MSI in the green and red edge are well positioned for deriving these indices. Results confirm the particular importance of the Sentinel-2 sensor for agricultural applications because it provides access to green and red-edge waveband data with high spatial resolution of 20 m.

Importantly, papers describing non-destructive N and Chl content retrieval employed spectral bands located far from the main red absorption band of Chl where absorption saturates at low-to-moderate Chl values. Use of either green or red-edge spectral regions makes it possible to avoid this saturation, retaining a high sensitivity to changes in Chl content (Gitelson, 2011a, Gitelson, 2011b, Hatfield et al., 2008, Ustin et al., 2009).

Collectively, the previously mentioned studies indicate that the use of remote sensing techniques may provide accurate measures of N content. This paradigm suggests that absorption by Chl provides the necessary link between remote sensing observations and canopy-state variables that are used as indicators of N status (Baret et al., 2007). However, previous studies dealt separately with N or Chl content estimation and, thus, cannot test the concept. To prove this concept, performance of remote sensing techniques to estimate both Chl and N contents in leaf and canopy should be investigated. This study evaluates performance of remote sensing techniques to estimate both N and Chl leaf and canopy contents in maize and, specifically, accuracy of N and Chl estimation using near future satellites Sentinel-2 and Sentinel-3. Firstly, we established relationships between N and Chl contents at the leaf and canopy levels and then tested the performance of Chl-related vegetation indices to retrieve N and Chl contents in both leaf and canopy scenarios. Secondly, we identified optimal spectral ranges in the green and the red-edge regions allowing accurate estimation of N and Chl contents in maize over a wide range of leaf area index values. Finally, we assessed accuracy of suggested technique using spectral bands of Sentinel-2, which allows monitoring of Chl and N contents in crops with high spatial and temporal resolutions.