Theses and Dissertations @ UNI

Availability

Thesis (UNI Access Only)

Abstract

The use of all-terrain vehicles (ATVs) is a relatively recent phenomenon, extended primarily across the last three decades in many Russian Arctic communities. One example of ATVs growth is the village of Teriberka, Kola Peninsula. This coastal community has only local roads, so all access to the interior utilizes all-terrain vehicles. These 4-wheel ATVs provide year-round mobility for multiple peoples and for multiple reasons. The use of heavy and powerful ATV vehicles in such vulnerable Tundra environments cause extensive damage to the vegetation that will last for a long time.

One of the objectives of this study was to detect and mapping all-terrain vehicle tracks by applying edge enhancement algorithms and Image classification techniques on high-resolution WorldView-2 imagery. The study also aimed to identify the spatial and morphological characteristics of the all-terrain vehicle tracks in Teriberka, Russia. Finally, the study attempted to analyze the impacts of ATV on different types of vegetation in different impacted and influences zones through calculating four vegetation indices (NDVI, MRESR, MSAVI2, and MCARI2).

The study revealed that the image enhancements techniques can detect all-terrain vehicle tracks with sufficient accuracy, while Object-based (9%) and Artificial Neural Network (14%) methods provided very poor results. Emboss NE and Laplacian can extract ATV tracks better than Sobel. The analysis of the tracks indicated that they total 39.52 km in length, yielding an average track density over the study area of roughly 0.23 km per square km; track density is not spatially uniform, there are large areas with no detectable tracks. The highest concentration of tracks was near the campsite and Gazprom site in the northeast. Lesser density areas exist around these sites and in the southwest of the study area. ATV drivers prefer to drive on low lying and gentle slope areas. Four different WorldView-2 derived vegetation indices; NDVI, MSAVI2, MCARI2, & MRESR were used to identify the vegetation condition at or near the ATV tracks and in ATV track buffer zones. Surprisingly, some indices found more greenery, less stressed and more chlorophyll contained vegetation near the track than away, while others indicated no significant difference between vegetation condition at or near the track and away of the track.

The conclusion of this study is that high-resolution optical satellite image is well suited for mapping and detecting all-terrain vehicle tracks and also for surveying the damage on vegetation caused by all-terrain vehicles in Teriberka, Russia.

Year of Submission

2019

Degree Name

Master of Arts

Department

Department of Geography

First Advisor

Andrey Petrov, Chair

Date Original

2019

Object Description

1 PDF (X, 153 pages)

Language

en

Off-Campus Download

Share

COinS