Hydroclimatic Scenarios Using Dendroclimatic, Historical and GCM-Based Records Over the Northwestern Great Plains
Abstract
Historically, the Canadian Prairies have been highly susceptible to extreme
drought and pluvial events, resulting in economic hardship. There are strong associations
between large-scale circulation patterns (i.e. teleconnections), such as the low frequency
Pacific Decadal Oscillation (PDO), and higher frequency El Niño-Southern Oscillation
(ENSO) and North Atlantic Oscillation/Arctic Oscillation (NAO/AO), and the interannual
to multi-decadal hydroclimatic variations over western North American. Archives
of the longest instrumental period of record (~100 to 150 years) are unlikely to capture
the full range of hydroclimatic variability (50-70 years) or enable full assessment of the
links between the large-scale circulation patterns and the regional moisture conditions.
Multi-centennial, seasonal paleoclimatic reconstructions were derived for the
northwestern Great Plains using tree-rings located along the eastern Rocky Mountains of
Alberta and Montana. The variability in these hydroclimatic reconstructions was
explored on time scales of decadal to multi-decadal to assess the coherence between
natural climate oscillations and the frequency, severity and duration of drought and
excess precipitation. The results suggest that drought events are often associated with the
positive phase of the PDO and increased ENSO variance. The association between
pluvial events and the phases of the PDO and ENSO are not as clearly defined as the
drought events.
Due to the non-stationarity of the climate system, these historical climate trends
and climate variability cannot be projected forward. However, water managers,
stakeholders and policy makers are concerned how anthropogenic climate change will
alter these natural variability patterns. Global Climate Models (GCMs) are the only credible tool to derive future climate scenarios. Scenarios of future climate change and
hydroclimatic variability were derived using ten GCMs and three emission scenarios,
with numerous runs, that were best able to replicate the 20th Century temporal and spatial
characteristics of the relevant teleconnection patterns and the hydroclimatic variability
recorded in our tree-ring network. The multi-model mean future PDO projections, for the
early half of the 21st century, show a weak shift towards more negative PDO-like
conditions; however, the GCMs were split between those showing a shift, often
significant, between the more negative and positive PDO-like conditions for all three
scenarios.
This thesis is the first to: explore the pre-instrumental relationship between
extreme events, as identified using a paleoclimatic moisture reconstruction, and the largescale
circulation patterns (PDO and ENSO); explicitly project the PDO as calculated by
Empirical Orthogonal Function Analysis of North Pacific sea surface temperature
residuals; and, develop future scenarios of variability that incorporate the PDO, ENSO
and NAO. This complementary analysis, of inferred, instrumental and modeled
hydroclimate change and variability, produced a long-term time series to provide
perspective on low-frequency natural climate variability and explore the potential future
impacts of greenhouse gas warming on the large-scale atmosphere-ocean circulation
linked with these hydroclimate oscillations. Risk assessments will help identify our
vulnerability, and increase our adaptive capacity, to projected changes in water quantity
and quality, introduction of new diseases (waterborne or airborne), insect infestations,
shifts to fire occurrence and some of the economic concerns associated with changes and
uncertainties to the hydroclimate.