Earth and Planetary Sciences ETDs

Publication Date

Summer 8-1-2023


Lengthy and well-dated terrestrial records of the Quaternary Period with the appropriate proxies show how the critical zone responds to environmental change. In southwestern North America, resolving the many effects of aridity through time provides crucial context for addressing contemporary climate issues. However, arid conditions are generally non-conducive to paleo-record preservation, causing knowledge gaps during relevant climatic episodes. An ~80 m-long sediment core (STL14) from Stoneman Lake contains a continuous archive of conditions from local to regional scale since the early Pleistocene (~1.3 Ma) and is, therefore, a unique and immensely valuable scientific resource. STL is situated in a montane forest of Ponderosa on the southwestern edge of the Colorado Plateau in central Arizona. The bowl-shaped depression is likely a sinkhole and, if so, is one of the largest identified at ~100 ± 35 million m3. It drains a 3.5 km2 watershed composed of basalt flows that cap Paleozoic limestones at depth. Tephrostratigraphy, paleomagnetism, radiocarbon dating, sedimentary facies analysis, and palynology indicate the local ecosystem is sensitive to glacial-to-millennial-scale climate variability and has unique responses to each of the more than 10 glacial cycles experienced at the lake. Periodic deep lake conditions after ~900 ka indicate an intensification of glacial climates following the mid-Pleistocene Transition. Mineralogical (XRD) and geochemical (XRF) analyses of catchment materials show eolian dusts are a substantial component of lake sediments and surrounding soils. In lacustrine stratigraphy of the last two glacial cycles (Marine Isotope Stages (MIS) 7 – 1, ~230 ka – 0), quartz and illite predominate despite the catchment’s basaltic bedrock. Time-series of Zr/Ti and Rb/Ti indicate that eolian dust input increases relative to local weathering products vi during glacial periods and early interglacials, a result of decreased erosion rates in the basin. End member (EM) modeling of lake sediment particle size distributions (PSD) from laser granulometry distinguishes and quantifies different populations of clastic sediments, each shaped by the weathering and transport histories of their source material, throughout the stratigraphy. Five EMs explain ~99% of the data set’s variance; two are eolian. EM-3 is a fine dust (13 μm mode; 28% of total clastics) likely emitted >100 km away. Its influx (= mass accumulation rate = dry density • sedimentation rate • fraction of bulk sediment) increases rapidly after major climatic transitions (MIS 6>5 and 2>1), remains high during interglacials (MIS 1, 5, and 7), and correlates with regional alluvial and fluvial activity in the southern Great Basin, Mojave, and Sonoran Deserts. Loessic EM-4 (38 μm mode; 19% of total clastics) is deposited at the highest rates during glacials (MIS 6, 4, and 2) and probably originates tens of km upwind from aggrading floodplains and piedmont surfaces in the Verde River Valley. Overall dust mass accumulation rate (DMAR) averages 0.47 g·cm-2·kyr-1 over the past ~230 kyr, ranging from 0.001 to 8.8, and is on average 1.6x higher during interglacial periods than glacials. Dust flux out of the desert regions of southwestern North America is most dependent on geomorphic responses to climate change that expose fine sediment, and is not significantly affected by windiness, gustiness, wet removal, or surface roughness. The divergent response of SW regional dust flux from other global records showing higher rates of dust accumulation during glacial periods shows this region lacks major influence from glaciogenic sources and is instead greatly affected by the responses of its semiarid alluvial and fluvial systems to cycles of warm-dry and cool-wet climates. EM-1&2&5 (53% of total clastics) is locally sourced; its modes (0.18, 5.3, and 169 μm) correspond to the chemical and physical weathering products of local basalt. Its influx parallels with TiXRF, an indicator of local erosion, and shows catchment basalt denuded at an average rate of at least 0.2 mm/kyr over the last two glacial cycles. Sediment flux generally increases with pollen abundance from the piñon-juniper woodland biome, fire activity, and aridity. Altogether, paleoenvironmental indicators, upland sediment fluxes at STL, and downslope fluvial aggradation inferred from loess accumulation illuminate the causes and pace of sediment transport between hillslopes, tributaries, and rivers on the southwestern margin of the Colorado Plateau. Upland soil mantles are produced during cooler and wetter glacials on 104-year timescales, stabilized by extensive forest canopies. Following transitions vii to warmer and drier interglacial conditions, pine and subalpine forests migrate, wildfires activate, and precipitation modes change, increasing runoff and liberating hillslope sediments on 102 -103-year timescales. Much of this sediment is stored in tributaries and does not reach the Verde River mainstem until tens of thousands of years later when discharge increases during the next glacial episode.

Degree Name

Earth and Planetary Sciences

Level of Degree


Department Name

Department of Earth and Planetary Sciences

First Committee Member (Chair)

Dr. Peter J. Fawcett

Second Committee Member

Dr. R. Scott Anderson

Third Committee Member

Dr. Leslie D. McFadden

Fourth Committee Member

Dr. Tyler Mackey

Fifth Committee Member

Dr. Yemane Asmerom




Quaternary, paleoclimate, southwestern US, geomorphology, dust, critical zone

Document Type


Available for download on Friday, August 01, 2025

Included in

Geology Commons