Using species accumulation curves to study change through time in a diverse butterfly fauna along an elevational gradient

The motivation for this thesis comes from ecological questions about the variability in the population dynamics of butterfly species across geographically diverse locations within Northern California. The goal of this thesis can be summarized in the following parts: i) to parameterize and fit the skewed log-logistic model to the observed species accumulation curves at each location; ii) to develop associations between the estimated parameters of the accumulation curves (response) and the weather variables (predictors) for each site, and iii) to analyze the fit of the models and interpret the findings in statistical and ecological terms. Ten locations were analyzed. Annual butterfly species data were available for the period from 1973 to 2016 with small site to site variation. Weather variables considered for the models were seasonal and annual precipitation totals and maximum/minimum seasonal temperatures. We found that a majority of inter-annual variation in weather was explained by variation in precipitation. Associations between the parameters of the species accumulation curves were modeled using linear and polynomial regression tools. Fit was assessed using the mean squared error. Models were developed using each SLL parameter as the response variable and the seasonal weather variables as the predictors. Through the use of step-wise regression model selection, an optimal model was developed from this initial model for each site analyzed. All computing work was done using R software.

Level thesis:degree_level

Master's Degree

Year dc:date.issued

2022

Author dc:creator

• Pacheco, Alejandro

Advisor dc:contributor.advisor

• Panorska, Anna

Committee members dc:contributor.committeemember

• Kozubowski, Tomasz
• Forister, Matthew
• Harris, Jr., Frederick C

Handle dc:identifier.uri

http://hdl.handle.net/11714/8700

OAI identifier oai:identifier

oai:scholarwolf.unr.edu:11714/8700