On the twenty-first-century wet season projections over the Southeastern United States
There is a growing demand among policymakers for regionally downscaled climate projections. Current global projections are on the order of hundreds of kilometers, which is too coarse to use for decision-making. Furthermore, low-resolution models may not be adequate for resolving local surface features. Failing to resolve significant features such as coastlines or orography may severely hinder global climate projections. As a response to the growing demand, many downscaled modeling products have been developed, such as the European ENSEMBLES project (Doblas-Reyes and Goodess 2005) or the North American Regional Climate Change Assessment Program (NARCCAP) (Mearns et al. 2009).
NARCCAP presents users with easily accessible, dynamically downscaled data of global climate model projections over North America. NARCCAP model integrations cover periods of 1971-2000 and 2041-2070 that are representative of the current and future climate. Mean twenty-first century summer rainfall projections vary depending on the pairing of GCM and RCM. For instance, one of the NARCCAP regional climate models (MM5I) projects a drying of the Southeast United States (SEUS) when forced with the global CCSM3 data. When the same regional model is forced with global HadCM3 data, the MM5I projects an increase in summer precipitation over the SEUS. NARCCAP offers eleven such pairings of global and regional models. Sobolowski and Pavelski (2011) used a reliability ensemble averaging technique to weight and determine model projections of precipitation in NARCCAP models. Weighted rainfall averages were seen to be heterogeneous in sign and magnitude across the domain. Only the area west of the Mississippi River was seen to be consistently drier in a future climate. Elguindi and Grundstein (2013) noted, however, that much of the SEUS would have a lower moisture index under the Thornthwaite climate classification system in a future climate.
Of particular relevance is the diurnal variability of rainfall, which accounts for a large fraction of the total seasonal summer rainfall variability (Carbone and Tuttle 2008; Bastola and Misra 2013). Local sea breeze effects along the long SEUS coastlines are a significant contributing factor to diurnal rainfall variability (Schwartz and Bosart 1979; Biggs and Graves 1962; Atkins et al. 1995; LeMone 1973). Sea breeze variability in the SEUS peaks in the summer (Zhang et al. 2009). Orographic effects in the central SEUS also play a major role in the diurnal variability of the rainfall (Parker and Ahijevych 2007) despite the absence of persistent low-level jets (Zhang et al. 2006). The observed, infrequent low-level jets are most pronounced near the Appalachian and Blue Ridge Mountains in summer. Short-lived, convective episodes off the Appalachians accounted for 77.9 % of the summer seasonal variance in the mountainous regions (Parker and Ahijevych 2007). El Niño Southern Oscillation (ENSO) can also contribute to the variability of rainfall in the SEUS (Kushnir et al. 2010) primarily through its noted influence on landfalling tropical cyclones (Bove et al. 1998). The influence of ENSO on summer rainfall variability, however, is decadally varying and spatially sporadic over Florida and its bordering states (Misra and Dinapoli 2012). However, the modification of the diurnal variability of rainfall from global climate change could be another potential source of significant rainfall change in the SEUS.
Rainfall variability in the southern part of the SEUS is driven in part by warming of SST (Rauscher et al. 2011) in remote tropical oceans (Xie et al. 2010). Warming of equatorial Pacific SST is expected as a consequence of increases in greenhouse gas (GHG) concentrations (Vecchi et al. 2008). This anomalously warm tropical SST will likely result in reduction in SEUS rainfall (Rauscher et al. 2011), a characteristic of the "upped-ante" (also known as the "rich-get-richer") mechanism (Neelin et al. 2003). In a troposphere warmed by El Niño (Yualeva et al. 1994) or trapping of longwave radiation by increased GHG concentrations, more moist static energy (MSE) is required to initiate convection (Chiang and Sobel 2002). The source of this MSE is primarily atmospheric boundary layer (ABL) moisture. The regions abundant in ABL moisture (e.g., intertropical convergence zones) will generally convect more vigorously under a warm troposphere. But, Neelin et al. (2003) indicated that regions on the margins of these convective zones (typical subtropical regions with prevailing large-scale subsidence) would convect less in such warm environments, as the boundary layer air in these regions usually originates from relatively drier areas.
The North Atlantic Subtropical High (NASH) has been identified as another source of rainfall variability in the SEUS (Li et al. 2011; Christensen et al. 2007). Li et al. (2011) indicated that because of global warming, the NASH has become stronger and shifted westward, with a concomitant increase in the interannual variability of summer rainfall over the SEUS in the last three decades. They suggest that with the systematic westward shift of the Bermuda high in the last three decades, its meridional movement has had an increased influence on the SEUS summer season rainfall leading to its increased interannual variability.