Clemson University
Pasture Biomass Estimation Using Crop Surface Modeling and Vegetation Indices
Abstract
dc:description.abstractBiomass estimations are a critical function of any grazing or haylage system. Determination of ideal harvesting or grazing times help optimize the quality and quantity of above-ground biomass (AGB). The objective of this study was to evaluate the use of unmanned aerial vehicles (UAVs) for Alfalfa and Tall Fescue biomass estimations. Using a DJI Mavic Pro, RGB and NDVI images were taken and used to create orthomosaic images. Structure-from-motion (SfM) techniques were used to developed digital elevation models to evaluate the change in canopy height (∆H) between pre- and post-harvests. Change in canopy height (∆H) was shown to provide the best estimation of predicted wet biomass (PBMwet) for both Tall Fescue (R2 = 0.88) and Alfalfa (R2 = 0.96). Predicted dry matter fraction (PDMF) functions were also created using a canopy density (CD) index, derived from NDVI, and maximum daily temperature (Tmax). Using the best PBMwet and PDMF functions, predicted dry biomass (PBMdry) models were created. For Alfalfa, the best PBMdry model utilized ∆H for PBMwet prediction and Tmax for PDMF predictions, producing an R2 of 0.955. For Tall Fescue, the best PBMdry model utilized ∆H for PBMwet prediction and CDfor PDMF predictions, producing an R2 of 0.865. Although these models produced the best results, other models produced similar results and may be better suited depending on resources availability, geographical regions, or access to historical weather data.
Degree
thesis:*- Name thesis:degree_name
- Master of Science (MS)
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Agricultural Education
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- MacInnis, Brendan
- Contributors dc:contributor
-
- Dr. Bulent Koc
- Dr. Matias Aguerre
- Dr. Aaron Turner
Subjects
dc:subject × 10Identifiers
dc:identifier.*- Repository record dc:identifier
- https://open.clemson.edu/all_theses/3734