Improved estimation of herbaceous crop aboveground biomass using UAV-derived crop height combined with vegetation indices

Abstract

Vegetation indices are used in precision agriculture to estimate crop aboveground biomass (AGB), and, in turn, to quantify crop needs. However, crop species and development stage affect vegetation indices limiting the setup of generalized models for AGB estimation. Some approaches to overcome this issue have combined vegetation indices and structural crop properties such as crop height. However, only a few studies have considered different herbaceous crops like forages and cover crops. A two-year field experiment was carried out on five winter cover crops with different habits at a high cover fraction (on average 93%) to study if combining vegetation indices, crop height and the fraction of soil covered by the crop could improve AGB estimation. Seven vegetation indices, crop height and cover fraction were derived from UAV-multispectral images. Species-specific and global (including all species) regression models were built and tested through cross-validation (CV). Green-based indices were the best estimators of AGB (RCV 2 = 0.56-0.93, normalized root mean square error in CV nRMSECV= 26-38%) of the five species, separately. A global linear model using crop height alone, provided good results (RCV 2 = 0.57, nRMSECV= 42%). Also, stepwise multiple regression was used to get a global model with crop height and five vegetation indices (RCV 2 = 0.75, nRMSECV= 31%). Finally, a model was proposed where AGB was estimated by a vegetation index until plants covered 97% of soil or its height was shorter than 125 mm, and by crop height for vegetation taller than 125 mm. The promising results (RCV 2 = 0.65, nRMSECV= 37%) suggested the possibility of increasing AGB estimation by considering both UAV-derived vegetation indices and structural crop properties.

Prediction accuracy and heritability of UAV based biomass estimation in wheat variety trials as affected by variable type, modelling strategy and sampling location

“…This study highlighted that combining geometric and spectral traits consistently led to improvements in prediction accuracy. This supports similar studies which found that sensor fusion can provide more comprehensive information surrounding canopy characteristics (Corti et al, 2023;Deery et al, 2020;Pauli et al, 2016) than individual sensors alone. At the same time, our comparison of spectral and geometric traits highlighted that geometric traits were more closely correlated with canopy biomass at all growth stages.…”

Assessing relationships of cover crop biomass and nitrogen content to multispectral imagery

“…Besides VI, the incorporation of satellite-derived radar (SAR) alongside red-edge NDVI in regression models improved biomass estimations, although they still plateaued at 1.9 Mg ha −1 (Jennewein et al, 2022). SAR is utilized in remote sensing to estimate plant heights, while drone-derived point clouds have enhanced stepwise linear regression models for crop height estimation (Corti et al, 2023;Kümmerer et al, 2023;Sangjan et al, 2022). Handheld cameras have also leveraged point clouds to improve biomass estimates for grass cover crops, with improvements of up to 8.0 Mg ha −1 reported (Dobbs et al, 2023).…”

“…Precision mapping of biomass offers an efficient solution to create field scale cover crop maps. Various sensing methods, including handheld, drone‐mounted, and satellite‐based sensors, have been employed for cover crop measurements (Corti et al., 2023; Prabhakara et al., 2015; White et al., 2019). These sensors have been used to measure biomass through vegetation indices (VIs), such as the normalized difference vegetation index (NDVI), triangular vegetation index (TVI), or Green NDVI (Prabhakara et al., 2015; Yuan et al., 2019).…”

“…The use of drones, or uncrewed aerial vehicles (UAV), is being explored across crop production (Gao & Zhang, 2021; Meresma et al., 2020; Poley & McDermid, 2020) and is expanding into cover crop measurements (Corti et al., 2023; Yuan et al., 2019). Drones provide higher resolution than satellites and can be flown on demand, depending on local weather.…”

Inversion of winter wheat leaf area index from UAV multispectral images: classical vs. deep learning approaches

“…These spectral bands and the vegetation indices have been effectively employed to predict crop LAI, demonstrating commendable accuracy ( Liu et al., 2021 ; Wang et al., 2020 ). However, limitations arise when estimating LAI solely using visible or multispectral vegetation indices, especially in the presence of high crop cover and saturation phenomena ( Corti et al., 2022 ; Zhou et al., 2017 ). To address these limitations, some researchers have incorporated texture features alongside vegetation indices for crop parameter estimation ( Liu et al., 2023b ; Xu et al., 2022 ).…”