UC Berkeley

The Atacama Desert is a coastal desert in South America spanning both Peru and Chile. While the desert experiences a range of climate regimes, perhaps the most unique region of the desert lies in the north of Chile where annual precipitation is less than 2 mm. This hyperarid region of the desert often goes decades without precipitation and is believed to have maintained a semi-continuous hyperarid climate for the last few million years.

Hyperaridity has had a profound impact on the geochemical and biological processes of the region. One of the most notable and economically important features of the desert are the salars. Salars are the location of past lakes that have lost their surface water due to changes in climate and/or tectonic activity. They provide minable concentrations of iodine, boron, and other salts, and are recognized as one of the only habitats for microbial life in this region. Presently, they are covered in thick and rugged salt crusts, often made of sodium chloride (halite) and calcium sulfate (gypsum).

Due to the extreme and persistent aridity, shallow groundwater is pulled upward towards the surface via capillary flow. This upward water movement is the reverse direction of most well drained desert soils, and leaves a diagnostic imprint in the soil chemistry. The first chapter of my dissertation explores how soils in salars differ from those found in other more humid deserts, and proposes changes to the USDA Soil Taxonomy that would allow for a more accurate and informative classification of them.

To further understand and catalogue the rates and processes of soil formation in this unique environment, in my second chapter I examine the chemical and isotopic profiles of two soils of differing ages in the Salar Llamara, Chile. I found that soil development is actively occurring, with evaporation of shallow groundwater driving the major geochemical processes. The upward movement of water produces a distinctive salt profile, with the most soluble salts concentrated on the surface and the least soluble salts near the base.

Through monitoring and experimental work, I calculated long-term evaporation rates, and found that they decrease with soil age as the salt crust thickness increases. While it has been suggested that these salars are part of an ancient landscape, this work provides evidence that they are dynamic and evolve on relatively short timescales. Finally, despite the lack of rainfall, I found that the crusts are able to sustain microbial growth by buffering environmental changes in temperature and relative humidity.

While the hyperarid region of the Atacama Desert only receives rare and infrequent precipitation, it is exposed to marine fog from the Pacific Ocean on a regular basis. The amount of fog changes predictably from west to east as the distance from the coast increases. For many years the Atacama Desert was suggested to be the dry limit to life. However, in the last decade it has been recognized that microbial communities are capable of surviving in salt crusts on the surface of salars. At a relative humidity over 75%, halite is able to absorb enough moisture from the atmosphere to create a saline solution in its mineral pores. In this way, salt crusts can provide organisms with liquid water in the absence of precipitation.

My third chapter is a detailed investigation of the microbial community composition and structure in this unique ecosystem. Using next metagenomics techniques, I reconstructed 124 distinct draft quality genomes from three sites along a fog frequency transect. All communities are comprised of a large variety of Halobacteriales, Salinibacter, Chlorophyta, and Cyanobacteria. Additionally, some communities contain lower abundances of Nanohaloarchaea, Actinobacteria, Gammaproteobacteria, Thermoplasmatales, and Naegleria. Candidate Phyla Radiation bacteria (OD1 and TM7) not previously reported from hypersaline environments before, were also found in low abundance in some of the communities.

While there is a lot of overlap in community membership across sites, samples cluster by site based on bacterial and archaeal abundance patterns. The concentration of photosynthetic organisms declines with increasing distance from the fog source (Pacific Ocean), and radiocarbon dating showed that carbon cycling is occurring more rapidly in sites with more fog events. I conclude that the strongest driver of community membership is fog delivered moisture, controlled by proximity to the coast.

The final chapter of my dissertation explores a unique late Quaternary paleoenvironmental record in loess deposits. Within what is largely a deflationary landscape, I found small loess dunes that have been slowly accreting for the last 20,000 years in depressions of the highly eroded Soledad formation. One of the unique local climate characteristics of this area is the strong and persistent on-shore winds.

Radiocarbon ages of organic matter embedded within the deposits show that the dunes began accumulating rapidly at the Last Glacial Maximum, and that the accumulation of sediment slowed considerably after the Pacific Ocean attained its present post-glacial level. Chemical and isotopic analysis of the sediment and fatty acids preserved within the dunes provide evidence for increased marine fog density and intensity of onshore westerly winds beginning 10,000 yr BP. At this time, grain size increases while accumulation rates simultaneously decrease, suggesting greater wind speeds and/or decrease in sediment supply. Organic sediment δ15N values steadily decrease, suggesting a shorter path length between N upwelling (Pacific Ocean) and N deposition in the dunes. This unique and continuous record of paleo-conditions provides a window into local processes occurring over the last 20,000 years.