Tennessee Valley surface water is important to economic and population growth in the southeastern United States. By expanding streamflow records, water planners and managers can make decisions based on hydrologic events not appearing in current instrumental records. In the following research, monthly flow data from six USGS streamflow gages on the Clinch, Emory, Holston, and Nolichucky Rivers is used to create seasonal and annual streamflow seasons. Approximately 70 tree-ring chronologies across the Southeast U.S. are prescreened by length and correlation analysis against 38 streamflow seasons revealing that the May-June-July (MJJ) streamflow period displays the best tree-ring climate signal. The screened chronologies are then entered into stepwise linear regression, and R2 values for the six models range from 0.36 to 0.52. Reconstruction models range indicate estimation errors due to multicollinearity of the streamflow and tree-ring chronology datasets are minimal. The Durbin-Watson statistics indicate the model residuals do not autocorrelate, except for the Nolichucky River streamflow model, which may possess serial correlation. The positive values of the RE parameter indicate each of the models have statistical skill, and the RMSE parameter provides error ranges equal to 18 to 44% of the average observed instrumental flows. Based on the results, three gages, the Nolichucky, NF Holston, and SF Holston, were deemed acceptable. These models represent the first statistically skillful streamflow reconstructions in the Tennessee Valley. The reconstructions range from 209 to 295 years in length ending in 1980 and extending as far back as 1686. Examination of the reconstructions shows extreme drought in the 1770s. The wettest periods occurred from the 1970s to the mid-2000s. Other severe drought events occurred in the 1700s, the 1840s, and the early 1910s proving current records do not provide full accounts of Tennessee Valley streamflow variability.

Tree rings have been used as a proxy in reconstructing streamflow in the western U.S. for many years, but few reconstructions have been attempted in the eastern United States. Clear limitations exist for streamflow reconstructions in the eastern U.S. compared to the western U.S., but value can be established as demonstrated in this research. The primary goal of this research was to reconstruct streamflow using data from five headwater gages in the Appalachian Mountains of North Carolina. These gages are located on the Valley River, the Oconaluftee River, the Nantahala River, the Little Tennessee River, and the Watauga River. Tree-ring chronologies were used to reconstruct streamflow. Tree-ring chronology predictors were selected using a seasonal correlation analysis. Seasonal correlation analysis revealed May-June-July (MJJ) streamflow variability being highly correlated with tree-ring chronologies in the study region and vicinity. Stepwise linear regression methods were used to reconstruct MJJ streamflow. The reconstructions for the Valley, Oconaluftee, and Nantahala Rivers were considered acceptable reconstructions because the models explained approximately 50% of the total variance in historic period MJJ streamflow records. These three streamflow reconstruction models have predictive skill indicated by a positive reduction of error (RE) values. The root mean square error (RMSE) statistic was 11.5 million cubic meters (MCM) for the Valley River (26% of the mean reconstructed MJJ flow), 15.9 MCM for the Oconaluftee River (16% of the mean reconstructed MJJ flow), and 8.2 MCM for the Nantahala River (20% of the mean reconstructed MJJ flow). Analysis of the reconstructed streamflow data for these three rivers revealed low flow periods from 1710 to 1712 at all three sites. The research presented here shows the potential benefit of using tree-ring chronologies to reconstruct streamflow in the Tennessee Valley region by demonstrating the ability of proxy-based reconstructions to provide useful data beyond the instrumental record. These useful data include identification of extreme wet or dry periods and oscillations in the historical reconstructions that are not visible in the instrumental data.

In the southwestern United States, the North American monsoon (NAM) is the main driver of severe weather and accounts for nearly half the annual precipitation. How the monsoon has behaved in the past and how it will change in the future is a question of major importance for natural resource management and infrastructural planning. In this dissertation, I present the results of three studies that have investigated North American monsoon variability and change from the perspective of paleoclimate records, future climate change projections, and simulation of the low-frequency variability with the longest retrospective atmospheric reanalysis. In the first study, a monsoon-sensitive network of tree-ring chronologies is evaluated within its ability to reproduce NAM variability during the past four centuries. Matrix methods are used to detect the low-frequency spatiotemporal variability. The treering chronologies can reasonable characterizes the dominant modes of NAM climate variability. The monsoon tree-ring network is able to reproduce the interannual variability of cool and warm season precipitation, in a manner similar to the period of the instrumental record. Earlywood and latewood adjusted chronologies reveal low frequency climate variability at decadal and longer timescales that is beyond the ability of the instrumental record to temporally well resolve. This low-frequency climate variability seems to be part of a much larger cycle that coincides with the occurrence of multiyear persistent droughts. In the second study, we consider the modes of natural climate variability identified in the previous study to objectively assess the degree of physical uncertainty in climate change projections for NAM from Regional Climate Models (RCMs) used in the North American Regional Climate Change Assessment Program (NARCCAP). Climate change projection models are evaluated mainly on their ability to represent warm season driven by quasi-stationary Rossby wave trains and El Niño Southern Oscillation – Pacific Decadal Variability (ENSO-PDV). It is concluded that use of the NARCCAP model ensemble mean for NAM climate projections is probably not suitable. NARCCAP RCMs are largely a slave to their driving global models and their error in the specification of large-scale atmospheric circulation. Only one out of eight NARCCAP RCMs has a reasonable representation of the seasonal cycle of monsoon precipitation and ENSOdriven interannual variability in both the 20th and 21st centuries. No decadal variability was observed in any of the NARCCAP RCMs. In the third study, the low-frequency drought signal found with tree-ring chronologies is further explored within the framework of a regional climate modeling. Version 2 of the Twentieth-Century Reanalysis (DD-20CR) is dynamically downscaled over a contiguous U.S.-Mexico domain. Statistic analysis of the DD-20CR suggests that the low-frequency drought signal in the Southwest is driven by atmospheric circulation changes on global to continental scales that affect precipitation in Central American as well. DD-20CR reproduces the spatial patterns of precipitation associated with climate variability at decadal and longer timescales in a manner that compares well with observational records and tree-ring chronologies. Low-frequency climate variability is therefore likely responsible for the multiyear persistent droughts in the last four centuries, as independently evaluated from the tree-ring monsoon-sensitive network.