The incorporation of autonomous computation and artificial intelligence (AI) technologies into smart agriculture concepts is becoming an expected scientific procedure. The airborne hyperspectral system with its vast area coverage, high spectral resolution, and varied narrow-band selection is an excellent tool for crop physiological characteristics and yield prediction. However, the extensive and redundant three-dimensional (3D) cube data processing and computation have made the popularization of this tool a challenging task. This research integrated two important open-sourced systems (R and Python) combined with automated hyperspectral narrowband vegetation index calculation and the state-of-the-art AI-based automated machine learning (AutoML) technology to estimate yield and biomass, based on three crop categories (spring wheat, pea and oat mixture, and spring barley with red clover) with multifunctional cultivation practices in northern Europe and Estonia. Our study showed the estimated capacity of the empirical AutoML regression model was significant. The best coefficient of determination (R²) and normalized root mean square error (NRMSE) for single variety planting wheat were 0.96 and 0.12 respectively; for mixed peas and oats, they were 0.76 and 0.18 in the booting to heading stage, while for mixed legumes and spring barley, they were 0.88 and 0.16 in the reproductive growth stages. In terms of straw mass estimation, R² was 0.96, 0.83, and 0.86, and NRMSE was 0.12, 0.24, and 0.33 respectively. This research contributes to, and confirms, the use of the AutoML framework in hyperspectral image analysis to increase implementation flexibility and reduce learning costs under a variety of agricultural resource conditions. It delivers expert yield and straw mass valuation two months in advance before harvest time for decision-makers. This study also highlights that the hyperspectral system provides economic and environmental benefits and will play a critical role in the construction of sustainable and intelligent agriculture techniques in the upcoming years.

Remote sensing is a valuable tool for surveying submerged aquatic vegetation (SAV) distribution patterns at extensive spatial and temporal scales. Only regular mapping over successive time periods (e.g., months, years) allows for a quantitative assessment of SAV loss or recolonization extent. Still, there are only a limited number of studies assessing temporal changes in SAV patterns. ESA Sentinel-2 (S2) has a high revisiting frequency permitting the multi-temporal assessment of SAV dynamics both seasonally and inter-annually. In the current study, a physics-based IDA (Image Data Analysis) model was used for the reconstruction of past SAV percent cover (%cover) patterns in the Baltic Sea coastal waters based on S2 archived images. First, we aimed at capturing and quantifying intra-annual spatiotemporal SAV dynamics happening during a growing season. Modeling results showed that significant changes took place in SAV %cover: the extent of low-cover (0–30% coverage) and intermediate-cover (30–70% coverage) areas decreased, while high-cover (70–100% coverage) areas increased during the growing period. Secondly, we also aimed at detecting SAV %cover spatiotemporal variations inter-annually (over the years 2016–2022). Inter-annual variability in %cover patterns was greater in the beginning of the vegetation period (May). The peak of the growing period (July/August) showed greater stability in the areal extent of the %cover classes.

A hyperspectral reflectance database was acquired for Baltic Sea submerged aquatic vegetation (SAV) and bare substrates by using Ramses (TriOS GmbH, Germany) radiometers capturing spectral data within the visible (VIS) and near-infrared (NIR) spectral ranges. The target samples included the most dominant and characteristic SAV species in the Baltic Sea, as well as several bare substrate types and beach cast communities. Target samples were measured within the 350–900 nm wavelength range under sunlight conditions without the water column influence. That is, samples were taken out of the water. Such a library is expected to provide insight into the spectral properties of various SAV species and substrates occurring in the coastal waters of the temperate geographic regions, facilitating the development of algorithms for differentiating and mapping various SAV communities. Additionally, measured reflectance spectra can be used as spectral end-members in physical models and classification algorithms for coastal vegetation mapping and quantification. The data are openly available in the PANGAEA online repository at https://doi.org/10.1594/PANGAEA.971518 (Vahtmäe et al., 2024).

A hyperspectral reflectance database was acquired for the Baltic Sea submerged aquatic vegetation (SAV) and bare substrates by using Ramses (TriOS) radiometers capturing the spectral data within the visible (VIS) and near infrared (NIR) spectral range. The target samples included the most dominant and characteristic SAV species in the Baltic Sea, as well as several bare substrate types and beach cast communities. Target samples were measured within the 350 to 900 nm wavelength range under sun light conditions without the water column influence i.e. samples were taken out of the water. Such library is expected to provide insight into the spectral properties of various SAV species and substrates occurring in the coastal waters of the temperate geographic regions facilitating development of algorithms for differentiation and mapping various SAV communities. Additionally, measured reflectance spectra can be used as spectral endmembers in physical models and classification algorithms for coastal vegetation mapping and quantification. Data are openly available at PANGAE online repository https://doi.pangaea.de/10.1594/PANGAEA.971518 (Vahtmäe et al., 2024).