Date of Graduation

Spring 2026

Degree

Master of Science in Geography and Geology

Department

School of Earth, Environment & Sustainability

Committee Chair

Tasnuba Jerin

Abstract

Sinkholes are characteristic enclosed depressions with internal drainage connecting the surface and subsurface karst hydrologic system. Their morphological evolution involves gradual subsidence of the ground surface and hazardous collapses. Therefore, understanding sinkhole morphometry and linked spatial distribution is critical for managing geohazards in karst watersheds. Sinkhole occurrence shows spatial variability based on watershed geomorphological attributes, surficial hydrological patterns, hydrogeological and geological settings. The aim of this research is to investigate morphometric properties and distribution of sinkholes in the carbonate karst-dominated lands of Mark Twain National Forest in Missouri, USA. The specific objectives were to (1) identify sinkhole clusters based on their morphometry; (2) define the morphometric characteristics and the evolutionary trajectory of the sinkhole clusters; (3) link the identified sinkhole clusters with geomorphic and watershed drainage properties. A sequence of geospatial and statistical analyses were conducted using intrinsic and extrinsic variables collected for sinkholes. K-means clustering method was used to develop sinkhole clusters. Clusters were then investigated using Classification & Regression Tree (CART) algorithm to analyze their characteristics. Multinomial Logistic Regression was used to select the most important geomorphic and drainage properties that shape sinkholes. The results identified four distinct morphometric clusters. Cluster-1 and 3 show strong spatial dependence within each cluster across the study area. Cluster-1 was identified as uvalas, which are large order-of-magnitude karst depressions with valley-like features. Cluster-3 includes medium-sized deep sinkholes, formed by cover collapses and subsequent degradation. Cluster-2 includes medium-sized shallow subsidence sinkholes, while Cluster-4 contains the smallest, circular solution sinkholes with uniform occurrence pattern and minimal management concerns. These four classes of sinkholes also provide insights into the evolutionary trajectory of sinkholes in the study area.

Keywords

sinkholes, morphometrics, spatial distribution, clusters, karst environment

Subject Categories

Geomorphology | Hydrology | Physical and Environmental Geography | Spatial Science

Copyright

© Marcell Tóth

Download

Open Access

Share

COinS