Award Date


Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Committee Member

Stephen Rowland

Second Committee Member

Joshua Bonde

Third Committee Member

Ganqing Jiang

Fourth Committee Member

Gary Haynes

Fifth Committee Member

Alan Simmons

Number of Pages



The purpose of this dissertation research is to examine the paleoecology of the Columbian mammoth (Mammuthus columbi) and the pygmy mammoth (M. exilis) on the individual, population, and community level, emphasizing factors internal to their ecosystem that may have contributed to their extinction. I focus specifically on mammoths from Southern Nevada and Southern California to add regional data to the complex and evolving study of North American mammoth paleoecology.

Chapter 2 documents the permitted excavation of a partial mammoth (Mammuthus sp.) between 2016-2018 from late Pleistocene groundwater-discharge deposits near Fairbanks Spring in the Amargosa Valley, Nevada. A preliminary radiocarbon date from an in situ mollusk shell indicates a calibrated age of 25,580 – 25,149 cal BP, indicating the mammoth remains are of Last Glacial Maximum age. Recovered elements from the mammoth include the proximal portions of two articulated tusks within the (incisive bones) alveoli of the skull, skull bones that were disarticulated from the tusks prior to burial, a partial right scapula, the atlas, axis, and several other partial cervical vertebrae. The sedimentology and stratigraphy of the Fairbanks Spring mammoth site suggests a depositional environment along the margins of a perennial, marl-producing marsh with seasonally dry edges and seasonal influence from spring channels or ephemeral washes.

Chapter 3 examines mammoth paleoecology at the population level, using mortality profile data to test the top-down trophic control hypothesis for Pleistocene extinction. For this study, a set of standard tooth measurement and identification standards for this analysis were compiled after extensive literature review. Fifty-eight Columbian mammoth molariform teeth were selected from the Gilcrease Cauldron Spring Paleontological Site (GCSPS), Las Vegas Formation, NV, to determine the ages of the animals in African elephant equivalent years for the purpose of constructing an age (a.k.a. mortality) profile. A Recalibrated radiocarbon dates of a subset of these teeth range from 14,336-20,567 rcy BP, around the onset of deglaciation. Ripple and Van Valkenburgh (2010) proposed that top-down trophic control of Pleistocene megaherbivore populations was a precursor to extinction and might have been secondarily enabled by predation pressure from Clovis hunters. Data generated from my study support this hypothesis. My results suggest that mammoth population sizes in the Las Vegas Valley during the Terminal Pleistocene were controlled from the top of the ecological pyramid down and were likely below carrying capacity. This trophic relationship may have made them vulnerable to extinction by an external catalyst. Mortality profiles from this study do not suggest that the GCSPS mammoth assemblage experienced lethal drought-stress or resource unavailability. Results from my study and others are consistent with top-down trophic controls on proboscideans surrounding the Last Glacial Maximum that may have made them susceptible to extinction at the end of the Pleistocene.

Chapter 4 takes a similar approach to Chapter 3, constructing mortality profiles for the assemblage of Columbian mammoth teeth from Rancho La Brea (RLB), CA. This study was an opportunity to supplement RLB mammoth mortality profile data with paleoecological studies of Pleistocene large carnivores from RLB to test the top-down trophic control hypothesis by Ripple and Van Valkenburgh (2010). Thirty-eight fossil mammoth molariform teeth were selected from La Brea Tar Pits and Museum for this study to determine the tooth assignment and percentage of wear. I observed selective mortality within the mature-adult age class, with a background “Type A” pattern in other age classes. This pattern could be reflective of mortality during the warmest months of the year and warmer overall years between 40,000-14,000 cal BP. In particular, this mortality was likely additive, considering presumably healthy, mature adults became mired within the asphalt, but not likely contributing to population instability. This mortality pattern could also be a result of sex-selective mortality of mature adult males due to either entrapment or intraspecific male combat. My data, corroborated by other studies, do not suggest mammoth population instability or decline at RLB due to resource stress during the interval of 40,000-14,000 cal BP, allowing me to refute the alternative hypothesis of bottom-up control. Although I cannot support the top-down trophic control hypothesis directly with this dataset, data from numerous studies of large carnivores at RLB paint a picture of a diverse and competitive predator guild during the late Pleistocene.

Chapter 5 examines mammoth paleoecology at the community level on the Northern Channel Islands of California. It has been suggested that body size played an important role in the behavioral isolation of M. columbi and M. exilis on the Northern Channel Islands of California; however this hypothesis has not previously been tested. Additionally, Channel Islands mammoth tooth morphological variation prompts questions about the timing, cause, and number of speciation events within this insular mammoth lineage. In this study, I test a hypothesis that body size played a role in niche partitioning between Channel Islands mammoth taxa. All teeth sampled for this study from the Natural History Museum of Los Angeles County were originally identified as M. exilis; however, several are within the size ranges observed for M. columbi. It is unclear if these teeth morphologically all belong to the same species, M. exilis. I used bulk (n=6) and serial (n=2) stable carbon and oxygen isotope data to interpret the ecological niche of this subset of Channel Islands mammoths. The δ13Csc values of tooth enamel from bulk sampling and mean serial sample values ranged from -13.07‰ to -6.8‰, VPDB (σ = 1.92). The calculated δ18OVSMOW values of tooth enamel ranged from 26.29‰ to 28.13‰, VSMOW (σ = 0.64). The calculated δ13Cdiet values of all the mammoth teeth sampled, excluding LPS2, suggest they ate a C3-dominated mixed feeder diet. The δ13Csc values from LPS2 were outliers within this assemblage of teeth, suggesting a mixed feeder diet with a higher concentration of C4 plants than the other sampled teeth in this study. The stable carbon isotope data from this study are consistent with reconstructed plant communities from the Channel Islands during the Pleistocene, and Channel Islands mammoths likely ate a much more C3-dominated diet than mammoths on mainland North America. The range of calculated drinking water δ18OWater values from this assemblage of teeth are unusually high (2.63‰ to 4.54‰, VSMOW), compared to the modern δ18O value for precipitation in Santa Barbara County, CA (-6‰, VSMOW) (Kendall and Coplen, 2001), which I interpret to reflect a dependence on surface waters sourced from condensed fog. During the Pleistocene, fewer fog events could have led to a corresponding decrease in conifer canopy cover on Santarosae, which in turn could have led to lower rates of fog drip, so the limited fog that Santarosae experienced would not be condensed and captured into surface waters as efficiently. In this way, Channel Islands mammoths may have needed to decrease their dependency on fog-sourced surface waters. Considering phylogenetic relationships, morphological variation, and gene flow within the continental North American Mammuthus metapopulation (See overview in Widga et al., 2017), it is likely that M. columbi and M. exilis of the Northern Channel Islands were not genetically or behaviorally isolated. Specifically, from this dataset, there does not seem to be any significant relationship between tooth size and inferred dietary niche.


Age [mortality] profiling; Amargosa Valley; California Channel Islands; Las Vegas Formation; Mammuthus; Rancho La Brea


Ecology and Evolutionary Biology | Environmental Sciences | Paleobiology | Paleontology

File Format


File Size

84.5 MB

Degree Grantor

University of Nevada, Las Vegas




IN COPYRIGHT. For more information about this rights statement, please visit