Southeastern Section - 73rd Annual Meeting - 2024

Paper No. 43-6
Presentation Time: 3:10 PM

LANDSLIDE SUSCEPTIBILITY AND TREE RING ECCENTRICITY ALONG CUMBERLAND PLATEAU, EAST TENNESSEE


PALMER, Megan1, NANDI, Arpita2 and MCSWEENEY, Robert2, (1)Department of Geosciences, East Tennessee State University, 1276 Gilbreath Dr., Johnson City, TN 37614, (2)Department of Geosciences, East Tennessee State University, 1276 Gilbreath Dr., Johnson city, TN 37614

Landslides are common along the rolling hills and escarpments of the Cumberland Plateau in East Tennessee and have impacted the roadway infrastructure, including State Route (SR) 116 in Anderson and Morgan counties, TN. Tennessee Department of Transportation (TDOT) continues to monitor and remediate unstable hillslopes along SR 116 for decades and will benefit from a landslide susceptibility map of the region. This study has predicted the locations of susceptible hillslopes using statistical models: forest-based classification and logistic regression in the New River Basin that encircles the SR 116. This study also recognized the tree growth ring patterns in landslide susceptible areas. One hundred and ten landslide locations were identified with Tennessee LiDAR data, and 50% were field verified. Geomorphic variables used in the model preparation included annual rainfall, distance from streams, elevation, slope angle, bedrock geology, landcover, soil type, and topographic curvature. Forest-based classification indicated annual rainfall was the most important factor in landslide susceptibility, followed by distance to stream and bedrock geology, with 84.5% correctness in the model’s performance. Logistic regression indicated all geomorphic variables except soil type, landcover, and curvature contributed to the landslide. The logistic regression model produced 73.6% overall correctness with a Receiver Operating Characteristic of 0.676. The study found areas of varying landslide susceptibility based on geomorphic factors. Additionally, from the landslide prone areas, nine bent trees were sampled through tree coring, the cores mounted, sanded, tree ring eccentricity was calculated and compared with the tree ring eccentricity from three trees in non-landslide areas. All nine tree rings showed moderate to high eccentricity, hinting that tree ring eccentricity can provide meaningful information about hillslope movement. However, more extensive research is needed to study the effect of landslides on tree growth patterns.