Using remote sensing indices to analyze the influence of bare ground in dust source areas to total dust-on-snow load in the san juan mountains, colorado

Abstract

The movement of dust across the western United States (US) has increased exponentially over the last 20 to 30 years driving a positive feedback regime altering the timing and magnitude of snowmelt. Dust radiative forcing of snowmelt can potentially exceed present day and likely future greenhouse gas forcing by two orders of magnitude. The semiarid landscape of the Colorado Plateau is one of the largest sources of dust in the western US. MODIS satellite imagery has been used to identify frequent, large-scale dust plumes that originate in the dust source area of northeastern Arizona, US and deposit that dust in the San Juan Mountains, Colorado. My study attempted to distinguish a "tele-link" between vegetation vigor in the dust source area and end of season total dust load in the San Juan Mountains from October 1 to June 30 for the years 2016 to 2021. The Normalized Difference Vegetation Index (NDVI) is a principal index tool used in multitemporal vegetation monitoring, and is commonly used as a direct indicator of vegetation health and growth. NDVI allows us to delineate the distribution of vegetation and bare soil based on the characteristic reflectance patterns of green vegetation. My study compared monthly NDVI mean values acquired by MODIS and Sentinel-2 to evaluate each satellites efficacy at modeling vegetation cover. Results suggest the association between vegetation vigor, bare soil, and total dust load is more complex and a number of factors could influence the inter-annual variability of dust-deposition. Statistical analysis employing ANOVA and multiple means comparison effectively identified pairwise groups who's monthly NDVI mean values were significantly different from others and 95% confidence intervals of the true expected difference, but failed to distinguish a "tele-link" between change in vegetation vigor and end of season total dust load. Finer-spatial resolution imagery captured more local variability in change in vegetation vigor over time and expanded the significant NDVI sampling window from 30 to 60 days. Projected climate change will likely increase aridity in the southwestern US, reduce the amount of vegetation cover, increase the amount of bare soil and enhance dust emission throughout the years.