An Analysis of Aquatic Vegetation in the San Marcos River Using sUAS
Date
2021-08
Authors
Lopez, Alexa
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Recreational usage is one of the largest impacts to the endemic Texas wild-rice (TWR) on the San Marcos river (SMR). During the summer of 2020, the restricted use of public access points to the river systems allowed a unique view of how aquatic vegetation responds with little intervention from river users. This thesis utilized small unmanned aerial surveillance (sUAS) to capture ultra-high resolution imagery of the SMR. The method used for classification featured an object-based image analysis using random forest, an AI algorithm to classify the data. The classified data had an accuracy assessment done which indicated an accuracy range 75.72% - 80.57% with a Kappa
range of .59 - .70. The classified imagery was then used in a change detection from July to August, August to September and July to September to determine the change in vegetation composition during the summer months. During the study period, there was expansion of stands that were identified as exclusively TWR, and the expansion of mixed vegetation stands, indicating the growth of the aquatic vegetation system. This study provided a continuous coverage of aquatic vegetation from a planar view using sUAS during a unique period where normally the river would have experienced its highest usage.
Description
Keywords
sUAS, Water resources, Remote sensing, GIS, Change detection, Classification, Random forests, River, Aquatic vegetation
Citation
Lopez, A. (2021). <i>An analysis of aquatic vegetation in the San Marcos River using sUAS</i> (Unpublished thesis). Texas State University, San Marcos, Texas.