Satellite eyes on alpine skies: A comparative study of modeled and remotely sensed vegetation indices using 21 years of field data.

Moore, Miles

Undergraduate Honors Thesis

Satellite eyes on alpine skies: A comparative study of modeled and remotely sensed vegetation indices using 21 years of field data. Public Deposited

Analytics

Download PDF

Citations

Citeable URL: https://scholar.colorado.edu/concern/undergraduate_honors_theses/ww72bc93p

Abstract

The Landsat (NASA & USGS) satellite data record has entered its fifth decade. Recent investigations leveraging Landsat surface reflectance data have revealed substantial greening trends in Normalized Differential Vegetation Index (NDVI) time series across the North American arctic tundra as well as in the alpine tundra of the European Alps. However, limited research has been conducted over the Southern Rocky Mountains (SRM) in this regard as these techniques often suffer from data scarcity in tundra regions thus complicating longitudinal trend analysis. In this study, I address this gap by evaluating three different models used to reconstruct spectral vegetation index (SVI) time series using 21 years of above ground biomass field data from the Niwot Ridge Long Term Ecological Research site. I propose a novel model that minimizes temporal autocorrelation while addressing the data scarcity problem and then, using Mann-Kendall tests, evaluate regional trends in SVI data over the SRM alpine tundra. My results reveal that greening and browning trend analyses are extremely sensitive to common modeling and SVI choices. Finally, I show that the SRM alpine tundra may be experiencing substantial browning trends – indicative of a systematic loss of biomass across almost a third of the region. These findings contrast the overwhelming greening trends identified in the tundra of the European Alps and North American arctic, carrying extremely important implications for the health and climate vulnerability of the SRM tundra.

Creator