Integrating Remote Sensing and Mid-Infrared Spectroscopy to Assess Land and Soil Degradation in Forest Ecosystems: Implications for Sustainable Management

A reliable assessment of forest resource stocks, productivity, and harvesting is a central goal of environmental monitoring programs. More specifically, evaluating appropriate management tools has become increasingly critical for assessing forest sustainability. Understanding how forests respond to various management tools is essential for developing and implementing sustainable strategies that enhance the resilience of forest ecosystems. Chestnut forest management practices differ across regions, with coppicing being one of the most common techniques. However, evidence from coppiced chestnut forests has raised concerns, particularly related to soil erosion and land degradation.

This study explores the use of advanced remote sensing and spectroscopic techniques to address two key aspects of land and soil degradation in Italian forests. The first objective is to utilize multi-temporal hyperspectral and multispectral satellite imagery to develop and test methods for identifying clearcut areas in chestnut forests resulting from coppice treatments, as opposed to other causes of bare soil, such as wildfires. The second objective focuses on monitoring the erosion impacts on land and soil degradation using mid-infrared (MIR) spectroscopy.

To achieve these goals, the study employed large-scale, multi-temporal satellite imagery from PRISMA, Sentinel-2, and Landsat 8, with a focus on developing a robust methodology for accurately delineating clearcut zones in chestnut forests located in central Italy (Campania). A pixel-based approach was used to differentiate between clearcut areas and pixels affected by other disturbances, beginning with a bare soil masking technique to create an annual bare soil composite image, followed by the delineation of clearcut zones.

In addition to remote sensing analysis, a comprehensive soil sampling campaign was conducted at active clearcut sites to evaluate the impact of chestnut management on soil degradation, with a focus on soil organic carbon content. Samples were collected from multiple locations within the clearcut areas to account for spatial variability. This dataset was used to identify areas vulnerable to soil erosion through MIR spectroscopy, offering valuable insights into soil function and the long-term impacts of management techniques on soil health.

The results show that the annual chestnut coppice clearcut areas were mapped with overall accuracies of 80%, 87%, and 92% for Landsat 8, PRISMA, and Sentinel-2, respectively. This approach enabled a detailed, high-resolution assessment of land use changes over time and the identification of clearcut zones due to coppice treatments. The use of MIR spectroscopy also facilitated the assessment and monitoring of erosion-prone areas within chestnut clearcuts.

The findings of this research have significant implications for forest management strategies, particularly regarding sustainable forest management and conservation. This study contributes to enhancing land management strategies by providing a deeper understanding of the environmental consequences of forest systems management techniques and highlighting the potential of remote sensing and spectroscopy for monitoring soil degradation.