The Potential of Using Spectral Reflectance Indices to Estimate Yield in Wheat Grown Under Reduced Irrigation

Abstract

The objectives of this research were to study the association in bread wheat between spectral reflectance indices (SRIs) and grain yield, estimate their heritability, and correlated response to selection (CR) for grain yield estimated from SRIs under reduced irrigation conditions. Reflectance was measured at three different growth stages (booting, heading and grainfilling) and five SRIs were calculated, namely normalized difference vegetation index (NDVI), simple ratio (SR), water index (WI), normalized water index-1 (NWI-1), and normalized water index-2 (NWI-2). Three field experiments were conducted (each with 30 advanced lines) in three different years. Two reduced irrigation environments were created: (1) one-irrigation level (pre-planting), and (2) two-irrigation level (pre-planting and at booting stage), both representing levels of reduced moisture. Maximum yield levels in the experimental zone were generally obtained with 4-6 irrigations. Genotypic variations for all SRIs were significant. Three NIR (near infrared radiation) based indices (WI, NWI-1, and NWI-2) gave the highest level of association (both phenotypic and genotypic) with grain yield under both reduced irrigation environments. Use of the mean SRI values averaged over growth stages and the progressive integration of SRIs from booting to grainfilling increased the capacity to explain variation among genotypes for yield under these reduced irrigation conditions. A higher level of broad-sense heritability was found with the two-irrigation environment (0.80) than with the oneirrigation environment (0.63). Overall, 50% to 75% of the 12.5% highest yielding genotypes, and 50% to 87% of the 25% highest yielding genotypes were selected when the NWI-2 index was applied as an indirect selection tool. Strong genetic correlations, moderate to high heritability, a correlated response for grain yield close to direct selection for grain yield, and a very high efficiency of selecting superior genotypes indicate the potential of using these three SRIs in breeding programs for selecting increased genetic gains in grain yield under reduced irrigation conditions.

Genetic Analysis of Indirect Selection for Winter Wheat Grain Yield Using Spectral Reflectance Indices

“…It was reported earlier that normalizing can improve the relationship over a ratio index (Tucker, 1979). Babar et al (2006aBabar et al ( , 2006b reported that the normalized index did not produce signifi cantly better results over the ratio index for predicting yield and biomass in spring wheat. This is also evident in our studies.…”

“…The water-based indices showed an advantage over the vegetation-based indices for selecting the higher-yielding genotypes compared to the widely used vegetation-based indices. Ball and Konzak (1993) demonstrated the effi ciency of normalized diff erence vegetation index in selecting higher-yielding genotypes under water nonlimiting conditions using two readings at the grain-fi lling stage, while Babar et al (2006b) reported the higher effi ciency of the near-infrared-based indices for grain yield selection under partially irrigated spring wheat environments. We have shown that the water-based indices are effi cient in selecting higher-yielding genotypes in winter Figure 2.…”

Proximal hyperspectral sensing and data analysis approaches for field-based plant phenomics

“…Many other studies have identified optimum wavebands for a given application by calculating narrow-band NDVI for all possible waveband combinations for a given hyperspectral sensor (Fu et al, 2014;Hansen and Schjoerring, 2003;Thenkabail et al, 2000;Thorp et al, 2004). Babar et al (2006) demonstrated several narrowband spectral reflectance indices that explained genetic variability in wheat biomass. Mistele and Schmidhalter (2008) measured spectral reflectance of maize canopies from four view angles and found the spectral reflectance indices were strongly correlated (0:57 6 r 2 6 0:91) with total nitrogen uptake and dry biomass weight.…”

Remote Sensing, Crop Yield Estimation and Agricultural Vulnerability Assessment: a Case of Southern Alberta

“…Numerous research efforts have been devoted to seeking a quantitative relation between remotely sensed spectral information and crop yields, and consequently obtaining a robust estimation and forecasting for agricultural productions [8,[20][21][22][23][24][25][26][27][28][29][30][31][32][33].…”

“…The first one is based on crop growth models, which incorporates remote sensing data into agrometeorological or bio-physiological models [23,18,[31][32][33]. For example, Doraiswamy et al (2003) implemented the real-time assessment of the magnitude and variation of crop condition parameters into the crop model called Erosion Productivity Impact Calculator (EPIC) [34].…”