Influence of climate teleconnections on precipitation and drought in the western United States

Drought exists over a much longer temporal scale than do most natural hazards, unfolding slowly over a series of months, years and even decades. This creates a unique opportunity for government agencies to plan for drought well ahead of time. Drought variability at various spatial and temporal scales must be understood in order for drought planning to be effective. As would be expected, numerous scientists have shown that spatial and temporal patterns of drought are often region specific and depend heavily on climatic processes that include teleconnections, which are preferred modes of low-frequency variability in the atmosphere and ocean. However, many unanswered questions remain.; In this dissertation, drought variability of various spatial and temporal scales is investigated as well as the teleconnections that control portions of that variability in the western United States. The four basic research questions are (1) what is the best way to define drought in the West for historical comparisons; (2) what are the independent influences of important teleconnections on precipitation and drought in the West; (3) what are teleconnective influences on sub-regional variability of drought in the West; and (4) how does the phase of the Pacific Decadal Oscillation impact precipitation and drought during years of neutral ENSO in the West?; With regard to the first question, the key finding is that percentiles of yearly water-year averages allow for direct comparisons of indices. This method shows that the 1896--1904 drought was the worst in modern Arizona history. For the second question, the key finding is that the main teleconnections associated with drought in the western Unified States explain only a small portion of the overall variance of drought, leaving much uncertainty regarding drought forecasts. For the third question, the key finding is that sub-regional variability in individual watersheds is greater than previously thought and must be accounted for when interpreting regional climate modeling data. Also, leading modes of drought variability are consistently located in the western United States and the cold phase of the PDO has a more consistent drought signal than does the warm phase. For the fourth question, the key finding is that the influence of the PDO on ENSO is greatest for low-index (neutral) events as opposed to the high-index (El Nino or La Nina) events as previously suggested. Long-lead forecasts during years of neutral ENSO may be more successful if they are slanted towards being drier (cold phase) or wetter (warm phase) than normal.