Abstract

Antarctic soils are formed by extreme climate, minimal biological activity, and diverse geological settings. This study analyzes 201 soil samples across four ice-free Antarctic Conservation Biogeographic Regions - North-west Antarctic Peninsula, Transantarctic Mountains, South Victoria Land, and North Victoria Land - to create a geochemical database (pH, EC, CEC, C and N, and ion concentrations) as well as to determine if there are consistent trends between geochemical properties and soil forming factors (SCORPAN). Despite variation in geochemical properties and SCORPAN covariates, some consistent trends emerged across all regions and individual Antarctic Conservation Biogeographic Regions. Using SCORPAN covariates and multiple linear regression models, soil geochemistry prediction maps were made for all regions (weight percent carbon, pH, and EC), the Transantarctic Mountains (Na, CEC, EC, pH), and South Victoria Land (C/N, weight percent carbon, CEC, and pH). Model performance varied across regions, with the highest being weight percent carbon in all regions (R2 = 0.44), Na in the Transantarctic Mountains (R2 = 0.68), and C/N in South Victoria Land (R2 = 0.83). Parent material and climate factors were important variables for the strongest prediction models. Despite some overfitting in the models, this study provides valuable insights for future soil prediction mapping in Antarctica, with broader implications for soil and microbial community research.

Degree

MS

College and Department

Geological Sciences; Computational, Mathematical, and Physical Sciences

Rights

https://lib.byu.edu/about/copyright/

Date Submitted

2024-12-19

Document Type

Thesis

Keywords

Antarctica, soil chemistry, multiple linear regression, soil forming factors, prediction mapping

Language

english

Share

COinS