Undergraduate Honors Theses

Thesis Defended

Spring 2017

Document Type

Thesis

Type of Thesis

Departmental Honors

Department

Anthropology

First Advisor

Matt Sponheimer

Second Advisor

Robin Bernstein

Third Advisor

William Bowman

Abstract

Abstract

Portable near infrared (NIR) spectroscopy is a fast, dependable, and cost- effective method for assessing nutritional chemistry in different plant species. Portable NIR can streamline analysis and investigation within ecological field settings, allowing researchers to evaluate significant nutritional variation for among populations and species of plants. The use of NIR in agriculture is common, but the application of portable NIR in fieldwork is still new in ecology. NIR in field settings has the potential to open new avenues of research, and can also save time and money when used in more traditional projects. To test the applicability of portable NIR spectroscopy in the field, I used nutritional data from South African savanna plant samples generated using wet chemistry, and NIR spectra of the same samples, to create regression models to predict plant nutritional properties from spectral data. I created regression models to predict crude fat, crude protein, and acid detergent fiber (ADF) of all plant organs (root, leaf, inflor, fruit, rhizome, and stem) and also specifically for leaf. My findings indicate that portable NIR can be used to predict crude protein and acid detergent fiber of South, African savanna plant samples. My findings also indicate that portable NIR is valuable for researchers in field settings, providing them with instant access to information.

Included in

Anthropology Commons

Share

COinS