H13E-0466 1340h
1917-2002 Covariability of Climate and Hydropower Production; Impact on Energy Transfers in the Western U.S.
Climate and climate variability have large effects on energy supply and consumption in ways that are becoming predictable. Climate forecasts are now routinely made for lead times from a week or so to as long as a year or more. As yet, however, this capability is not widely used in power management. California's residential electrical consumption has regular peaks in winter and summer, while energy consumption in the Pacific Northwest has a strong winter peak. Seasonal to inter annual climate along the west coast varies in such a way that California and the Pacific Northwest are often out of phase. For instance, warm and dry winters in the Pacific Northwest (and hence reduced hydropower production) often occur at the same time as cool and wet conditions in California. These conditions can now be forecast with some accuracy as much as a year in advance. We have developed models that make use of climate forecasts to simulate both the hydrology and operations of the Columbia River, and the Sacramento-San Joaquin (combined State Water Project and federal Central Valley Project) reservoir systems. These models are driven by gridded daily historical climate data, now complete from 1916 to current. We have used these sequences of historical climate data to address the following questions: 1) If the current system of hydropower reservoirs, and thermal energy plants in both the Pacific Northwest (PNW) and California had existed throughout the period 1916 to present, how frequently would winter and summer power production and demand have been out of phase? 2) What is the potential for use of weather and climate forecasts with lead times from a few weeks out to a year or more for improving joint operation of Pacific Northwest and California energy generation, considering specifically the potential for incorporation of weather and climate forecasts in projection of both supply (hydropower) and energy demand? We expand on a preliminary analysis using the retrospective simulations to examine the potential for alternative operation of the existing power intertie between the Pacific Northwest and California (capacity 8000 MW) that might exploit out of phase climate behavior between the PNW and California.
H13E-0467 1340h
Relationships between Climate Variability, Drought and Model Soil Moisture
This research investigates the interannual variability of soil moisture as related to large scale climate variability. A three layer hydrological model VIC -3L (Variable Infiltration Capacity Model - 3 layers) was used in the Colorado River Basin and Mississippi River Basin over a 50 year period. The simulation focuses on the soil moisture generation and simulation have been developed between January 1950 and December 2002 at daily time step. Simulation was performed on 1/8 degree resolution for the water balance model of both basins and shows the interannual variability of deep soil moisture. Using wavelet analysis, deep soil moisture is compared to the Palmer Drought Severity Index (PDSI), precipitation, and streamflow to determine whether deep soil moisture is an indicator of climate extremes. Wavelet and coherency analysis for the Colorado River Basin and the upper Mississippi River basin indicate a strong relationship between droughts and the deep soil moisture. Relationships between soil moisture and other large scale climate variability (e.g., ENSO, PDO, sea surface temperatures) are also evaluated.
H13E-0468 1340h
Hydrologic Impacts of the Pacific Decadal Oscillation on ENSO Influenced Regions
Several regions in the United States [Pacific Northwest (PNW), Southwest (SW), North Central (NC), Northeast (NE) and Gulf of Mexico (GM)] display a strong El Niño-Southern Oscillation (ENSO) signal in which monthly precipitation (and resulting streamflow) vary in the winter-spring season following the onset of a summer season warm (El Niño) or cold (La Niña) phase of ENSO. Additionally, numerous studies have shown that in several of these regions (PNW, SW and GM), a distinct lag exists between the occurrence of a summer El Niño (or La Niña) and the resulting winter-spring streamflow variation. The concern for water planners is "how likely (probability) will winter-spring streamflow be above (or below) median if a summer season ENSO event occurs?" and, "can this probability be provided prior to the beginning of the water year (01 October)?" Research has shown that the Pacific Decadal Oscillation (PDO) can increase (or reduce) the effects of ENSO. PDO tends to oscillate with a characteristic period on the order of 50 years (a particular phase of the PDO will typically persist for about 25 years). This period differs greatly from ENSO events, which typically persist for 6 to 18 months. Water-year (October to September) average values of the PDO were examined for the period of 1941 to 2001. During this period, three distinct phases were observed: 1941 to 1961 (21 years - Declining), 1962 to 1987 (26 years - Rising) and 1988 to 2001 (14 years - Declining). Currently, we are in a Declining PDO phase, which should persist for approximately 10 more years. An additional concern for water planners is "will the current Declining PDO phase enhance or dampen a summer ENSO event?" Data was obtained for unimpaired streamflow stations in the PNW, SW and GM basins from 1942 to 2002 (61 years of data). Preliminary results in the SW basin reveal that the current Declining PDO phase does result in a higher likelihood of below median winter-spring streamflow, as a result of a summer La Niña event (based on the Southern Oscillation Index), when compared to the Rising PDO phase.
H13E-0469 1340h
Spatial and temporal characteristics of US drought, 1950-2000
Drought accounts for a large portion of economic and environmental losses from natural hazards in the US. It generally develops slowly over long time scales (seasonal to annual), may have large spatial coverage and exhibit long-term persistence that ensures that its impacts can be widespread and long-felt. Drought can be highly variable in space due to variability in climate forcings and perhaps more importantly through smaller-scale land surface heterogeneities. Thus monitoring and prediction can present considerable challenges, notwithstanding the practical difficulties in observing soil moisture over large scales. This study investigates some of the issues related to the spatial and temporal characteristics of drought over the US. We use as the basis of this study a drought index based on soil moisture data derived from a long-term (1950-1999), high resolution (1/8th degree) simulation of the land surface water budget over the conterminous US using the Variable Infiltration Capacity (VIC) model. Comparison of the occurrence of drought at regional scales shows distinct coherence between certain neighboring regions but this varies in time. We explore this in further detail to investigate issues such as the scale at which coherence breaks down and how coherence and variability change across the country (with respect to climate, topography, and vegetation). The analysis also reveals high persistence in drought occurrence (and wet spells) across the western US. In contrast the eastern US shows high temporal variability as dictated by shorter-term climate variability. Persistent signals in soil moisture and drought have been linked to large-scale climate oscillations and we investigate this through the analysis of teleconnections. These findings are also dependent on the soil horizon in question, as soil moisture at different depths varies over different time scales and we look at how spatial variability and temporal persistence varies through the soil profile.
H13E-0470 1340h
North American Droughts of the Mid-late Nineteenth Century: a History and Model Simulation
Unlike the major droughts of the 20th century that are readily identified in the instrumental record, similar events in the 19th century have be identified using a combination of proxy data, historical accounts and a sparse collection of early instrumental records. In the United States, three distinct periods of widespread and persistent drought stand out in these records for the latter half of the 19th century: 1856-1865, 1870 -1878 and 1890 - 1896. Each of these events is shown to coincide with the existence of an anomalously cool, La Nina -like tropical Pacific. To examine the physical mechanisms behind these droughts two simple experiments were performed: the first forces an atmosphere general circulation model (AGCM) with the observed history of SSTs everywhere from 1856 to 2001 (the GOGA experiment), the second forces the AGCM only with tropical Pacific SSTs, being coupled to a two-layer entraining mixed layer (ML) ocean (the POGA-ML experiment). The AGCM used is the National Center for Atmospheric Research (NCAR) Climate Community Model 3 (CCM3). The extent to which the ensemble mean of the climate model is able to simulate the observed drought intervals is a test of both whether the dry conditions are SST-forced, globally or from the tropical Pacific only, and also whether the model is able to capture this. Due to a sparsity of instrumental precipitation data at this time, proxy evidence from tree rings is used as verification. A comparison of modeled soil moisture with PDSI data, a proxy for soil moisture, from the North American Drought Atlas (Cook et al., 2001) is made. Both the POGA-ML and GOGA ensemble means capture the three multi-year droughts of the mid-late nineteenth century identified by the `drought atlas' PDSI data. In each case the models reproduce the large-scale features of these droughts, albeit with some error, indicating that the droughts were SST forced. The similarity of the POGA-ML and GOGA simulations implies that the component of the each drought signal that is forced by the SST is driven by the SST anomalies from within the tropical Pacific - a La Nina-like tropical Pacific. The global atmosphere-ocean context of the 1856 - 1865 `Civil War drought interval is also examined in more detail to elucidate the dynamical mechanism behind this particular drought - the severest of the century .