L’utilizzo della spettroscopia per la rapida determinazione delle proprietà chimiche e fisiche del suolo

Soil is currently defined as a “natural independent body,” an open, dynamic, and multifunctional system, tightly integrated with the natural environment, capable to provide goods and services for ecosystems and human well-being. As a result, soil plays a crucial role in achieving 8 to 17 of the Sustainable Development Goals. Specifically, soil is estimated to contribute 98.8% to global food production, store 72% of global freshwater resources, and sequester between 2 and 5 gigatons of CO₂ annually, making it fundamental for climate change mitigation. However, between 20% and 40% of the world’s land is degraded and their functions severely compromised, mainly due to climate change, land conversion, and intensified land use. Given soil's importance, international agreements are encouraging countries to establish soil monitoring plans that promote the adoption of soil conservation practices. A key step in this process is identifying sensitive indicators to represent land degradation and developing efficient tools to rapidly and accurately measure them. In the last decades, soil organic carbon (SOC) has emerged as a key indicator for monitoring global land degradation. In particular, water-extractable organic carbon (WEOC) has proven to be a sensitive indicator of SOC turnover and soil management practices. Similarly, SOC fractions measured via high-temperature catalytic combustion provide crucial insights. The role of physical fractions in regulating SOC and nutrient availability is also well-known, making it important to consider soil texture and its spatial variability alongside SOC concentration. Recently, spectroscopy has gained prominence as a rapid, sustainable, non-destructive, and reproducible method for determining several soil properties. Gamma-ray spectroscopy, for example, proved to be particularly effective in assessing sand, silt, and clay content and their spatial variability. However, its use is still limited to in-situ application, and its predictive accuracy is influenced by the concentration of the target variables. Additionally, mapping the distribution of soil properties is heavily dependent on the sampling strategy and the application of geostatistical methods. In contrast, diffuse reflectance spectroscopy can be applied both in the lab and in the field, yielding promising results in both proximal and remote sensing. This technique can measure multiple properties—such as sand, silt, clay content, SOC, and its labile fractions—in a single measurement. Although current applications of diffuse reflectance spectroscopy largely rely on advanced statistical models based on soil spectral libraries, hyperspectral imaging spectroscopy is considered the most promising application. UAV-based imaging spectroscopy, in particular, can generate high-resolution maps of topsoil variability. Despite the rapid data acquisition, pre-processing remains challenging, requiring specific expertise, significant computational power, and large memory storage, which are not yet widely available. After more than 30 years of research, the future of soil spectroscopy will depend largely on improving its accessibility for end-users. In this context, the development of a global spectral-based model capable of integrating and elaborate local spectral data from various platforms will be crucial for future advancements.