Forests present uniquely unstructured and dynamic environments where traditional mapping systems often struggle to perform reliably. This chapter contributes to the development of digital forests by conducting a comparative field study of modern LiDAR-based odometry techniques for 3D mapping. Focusing on recent open-source methods for LiDAR Odometry and Mapping (LOAM) and LiDAR-Inertial Odometry (LIO), we begin by identifying and analyzing a range of scan-matching algorithms relevant to natural environments. A subset of these techniques is selected through preliminary evaluations, and real-world datasets are collected using two distinct 3D LiDAR sensors. Field experiments are conducted in forested terrain to assess the performance of each method under natural conditions. The study includes data acquisition using a quadruped mobile robotic platform, enabling consistent and repeatable testing in outdoor environments. The comparative analysis focuses on mapping and localization accuracy, computational efficiency, and sensor performance. The results provide critical insights into the strengths and limitations of current LiDAR-based odometry approaches, particularly in the context of forest mapping. This work lays the groundwork for more robust perception systems in forestry robotics and contributes valuable field data and evaluation protocols for future research in digital forest technologies.