Stable crop yields require an appropriate supply of essential soil nutrients such as nitrogen (N), phosphorus (P), and potassium (K) based on the accurate diagnosis of soil nutrient status. Traditional laboratory analysis of soil nutrients is often complicated and time-consuming and does not provide real-time nutrient status. Visible–near-infrared (Vis-NIR) spectroscopy has emerged as a non-destructive and rapid method for estimating soil nutrient levels. Vis-NIR spectra reflect sample characteristics as the peak intensities; however, they are often affected by various artifacts and complex variables. Since Vis-NIR spectroscopy does not directly measure nutrient levels in soil, improving estimation accuracy is essential. For spectral preprocessing, the most important aspect is to develop an appropriate preprocessing strategy based on the characteristics of the data and identify artifacts such as noise, baseline drift, and scatter in the spectral data. Machine learning-based modeling techniques such as partial least-squares regression (PLSR) and support vector machine regression (SVMR) enhance estimation accuracy by capturing complex patterns of spectral data. Therefore, this review focuses on the use of Vis-NIR spectroscopy for evaluating soil properties including soil water content, organic carbon (C), and nutrients and explores its potential for real-time field application through spectral preprocessing and machine learning algorithms. Vis-NIR spectroscopy combined with machine learning is expected to enable more efficient and site-specific nutrient management, thereby contributing to sustainable agricultural practices.

Improving nutrient use efficiency and minimizing environmental pollution from excessive fertilization require appropriate nutrient management supported by continuous monitoring of soil nutrient levels during crop growth. As only a few real-time sensors for the measurement of soil nutrients are available, this study evaluated the potential of electrical conductivity (EC) sensors, which reflect the ionic concentrations of the soil solution, for real-time estimation of plant-available nutrient levels. Nitrogen and potassium were sequentially supplied to achieve cumulative application rates of 25–300% of the nutrient uptake-based fertilization rate. The relationship between cumulative fertilization rate and accumulated sensor-based EC increase was described using linear, polynomial, and nonlinear saturation models. Sensor EC increased linearly from 25 to 125% of the nutrient uptake-based fertilization rate, while higher application rates were better explained by the nonlinear saturation equation. Sensor-based EC showed strong correlation with soil ammonium nitrogen (NH₄⁺-N), indicating that the sensor effectively reflected nutrient dynamics. In open-field pepper soil, fertigation-induced increases in sensor EC followed the patterns predicted by both the linear and nonlinear saturation models established in the laboratory. These results demonstrate that EC sensors can be used for real-time monitoring of soil nutrient levels and may contribute to efficient nutrient management in open-field cultivation.