Wheat yield and protein estimation with handheld‐ and UAV‐based reflectance measurements. Issue 4 (27th September 2022)

• Record Type: Journal Article
• Authors: Walsh, Olga S.; Marshall, Juliet; Jackson, Chad; Nambi, Eva; Shafian, Sanaz; Jayawardena, Dileepa M.; Lamichhane, Ritika; Owusu Ansah, Emmanuella; McClintick‐Chess, Jordan R.
• Is Part Of: Agrosystems, geosciences & environment. Volume 5:Issue 4(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-09-27
• Subjects: Agriculture -- Periodicals; Agriculture -- Environmental aspects -- Periodicals; Soil science -- Periodicals; Food science -- Periodicals; Food science; Agriculture; Agriculture -- Environmental aspects; Electronic journals; Periodicals; 630
• DOI: 10.1002/agg2.20309
• Languages: English
• ISSNs: 2639-6696

Abstract

Abstract: Precision agriculture provides efficient means of obtaining real‐time data to guide nitrogen (N) management based on predicted crop profitability. This study was conducted to assess the efficacy of using in‐season measurements (plant height, biomass weight, biomass N, soil plant analysis development [SPAD], GreenSeeker [GS] normalized difference vegetative index [NDVI], and unmanned aerial vehicle [UAV] NDVI) at Feekes 5 (tillering) and Feekes 10 (anthesis) to estimate wheat ( Triticum aestivum L.) yield and protein. The secondary aim was to determine whether the accuracy of yield and protein prediction varies by wheat class and cultivar. Six cultivars—hard red spring (HRS) wheat 'Jefferson' and 'SY Basalt', hard white spring (HWS) wheat 'Dayn' and 'UI Platinum', and soft white spring (SWS) wheat 'Seahawk' and 'UI Stone'—were planted at two locations in Idaho in 2018–2020. Plots were arranged in a randomized complete block design with four replications with each cultivar evaluated at seven N rates (0, 50, 100, 150, 200, 250, and 300 kg N ha –1 ). The determination of the Pearson correlation coefficients revealed that all parameters were linearly correlated with yield except for SPAD at Feekes 5 and biomass weight at Feekes 10. Although estimation of in‐season grain protein remains a challenge, NDVI was strongly correlated with yield especially at Feekes 5. The accuracy of yield prediction was similar for all wheat classes. Comparable accuracy of yield estimation was achieved with GS NDVI and UAV NDVI. Both hand‐held and aerial‐based spectral measurements could be used to prescribe N rates to be applied during tiller formation when wheat yield can be optimized. Core Ideas: GreenSeeker (GS) and UAV NDVI at Feekes 5 shows the most potential for in‐season wheat yield estimation. Comparable accuracy of yield estimation was achieved with GS NDVI and UAV NDVI. The accuracy of yield estimation was comparable for all wheat classes and cultivars. Estimation of in‐season wheat protein content remains a challenge.