Academic Level at Time of Presentation
Graduate
Major
Earth and Environmental Sciences: Watershed Science
List all Project Mentors & Advisor(s)
Katharine Loughney, PhD
Presentation Format
Poster Presentation
Abstract/Description
UAS vs. In-situ Measurement Techniques for Stream Substrates
Key words: Geomorphic, Fluvial, ArcGIS Pro, UAS
Traditionally, geomorphic assessments of streams and fluvial structures have been conducted using in-situ field techniques. These techniques include taking Wolman (D50) pebble counts to characterize substrate, measurements of stream dimensions with tape, meter sticks, and Jacob staff. We studied the efficacy of using an uncrewed aerial system (UAS) to measure the substrate composition of a section of stream along the east fork of the Clarks River in Calloway County, Kentucky. The section consists of a 30-meter riffle-pool-run sequence, with variation in substrate size to test particle size detection with UAS imagery.
Stream substrate in this project was first assessed in-situ with traditional field methods. Wolman(D50) pebble counts measure the median particle size of the substrate in a stream section. Wolman pebble counts are accomplished by randomly taking 100 pebble samples using a gravelometer measuring grain sizes from sand up to cobble (diameter measured in millimeters). Substrate composition of a stream can also be assessed using aerial imagery and utilizing Object-based image analysis (OBIA) tools in ArcGIS Pro. The same stream section was then imaged using a DJI drone flown 5 meters above the stream bed. These images were combined through a mosaic process and rendered into a 3D model in Pix4D and then uploaded to ArcGIS Pro. The Deep Learning Image Analyst extension in ArcGIS Pro was used to measure particle size distribution. The median pebble count from the hand-measured assessment was compared to the median particle size detected through ArcGIS Pro OBIA. The total counts and cumulative frequency distributions for both data sets were statistically compared with one another via regression analysis to establish the degree of precision of data collection between the two methods.
We hope to assess precision between the two data collection methods. This study opens the door to future research using UAS to assess fluvial features and to complete geomorphic assessments. UAS and other remote sensors can be advantageous because they allow the researcher to assess areas that may be difficult to access and the larger areas can be accessed in less time, without compromising accuracy of data collection.
Fall Scholars Week 2025
Earth and Environmental Sciences Poster Session
Included in
UAS vs. In-situ Measurement Techniques for Stream Substrates
UAS vs. In-situ Measurement Techniques for Stream Substrates
Key words: Geomorphic, Fluvial, ArcGIS Pro, UAS
Traditionally, geomorphic assessments of streams and fluvial structures have been conducted using in-situ field techniques. These techniques include taking Wolman (D50) pebble counts to characterize substrate, measurements of stream dimensions with tape, meter sticks, and Jacob staff. We studied the efficacy of using an uncrewed aerial system (UAS) to measure the substrate composition of a section of stream along the east fork of the Clarks River in Calloway County, Kentucky. The section consists of a 30-meter riffle-pool-run sequence, with variation in substrate size to test particle size detection with UAS imagery.
Stream substrate in this project was first assessed in-situ with traditional field methods. Wolman(D50) pebble counts measure the median particle size of the substrate in a stream section. Wolman pebble counts are accomplished by randomly taking 100 pebble samples using a gravelometer measuring grain sizes from sand up to cobble (diameter measured in millimeters). Substrate composition of a stream can also be assessed using aerial imagery and utilizing Object-based image analysis (OBIA) tools in ArcGIS Pro. The same stream section was then imaged using a DJI drone flown 5 meters above the stream bed. These images were combined through a mosaic process and rendered into a 3D model in Pix4D and then uploaded to ArcGIS Pro. The Deep Learning Image Analyst extension in ArcGIS Pro was used to measure particle size distribution. The median pebble count from the hand-measured assessment was compared to the median particle size detected through ArcGIS Pro OBIA. The total counts and cumulative frequency distributions for both data sets were statistically compared with one another via regression analysis to establish the degree of precision of data collection between the two methods.
We hope to assess precision between the two data collection methods. This study opens the door to future research using UAS to assess fluvial features and to complete geomorphic assessments. UAS and other remote sensors can be advantageous because they allow the researcher to assess areas that may be difficult to access and the larger areas can be accessed in less time, without compromising accuracy of data collection.