Climate and Energy Action Plan (CEAP)

CLIMATE TRENDS & PRO JECTIONS | CITY OF ASHLAND, OREGON | FINAL REPORT | 34 end of the century and results in atmospheric carbon dio xide concentrations greater than 900 ppm that will continue to rise beyond 2100. See Figure 25 for a graphic representation of these differences. It is important to note that RCP2.6, which attains negative greenhouse gas emissions by 2100, is the only RCP scenario to keep global temperature likely below 2°C (IPCC, 2013). Figure 25. Carbon emissions and atmospheric carbon dioxide concentrations for RCP scenarios (Walsh et al., 2014a) . In an integrated climate -hydrology -vegetation modeling project called “Inte grated Scenarios of the Future Northwest Environment,” the coarse resolution (100 -300 km) of the CMIP5 GCM output was downscaled over the Western United States to a resolution of about 6 km using the Multivariate Adaptive Co nstructed Analogs (MACA) method (Abatzoglou & Brown, 2012) . The MACA approach utilizes a gridded training observation datase t to accomplish the downscaling by applying bias -corrections and spatial patt ern matching of observed large -scale to small -scale statistical relationships. The downscaled climate data was then used as an input to hydrology and vegetation models. Simulatio ns of historical and future climate for the 6 -km grid cell containing the city of Ashland were obtained at the daily ti me step for th e maximum temperature, minimum temperature , and precipitation variables from 1950 to 2099 for 18 CMIP5 GCMs and the two ava ilable RCPs (i.e., RCP4.5 and RCP8.5). The selected temperature and precipitation metrics were derived from these variables ( see Table 3). Streamflow and snow dynamics within the Integrated Scenarios project were simulated using t he Variable - Infiltration Capacity hydrological model (VIC version 4.1.2.1; (Liang, Lettenmaier, Wood, & Burges, 1994) and update s) run on a 6 km grid. Simulations of streamflow and snow water equivalent ( SWE ) are only available for 10 GCMs used as input s to VIC. Future projections for routed streamflow at sites in the Rogue Basin do not yet exist, but are being generated and are anticipated to become available in fall 2016. However, projections of runoff —the amount of water at a particular location before it flows into a stream —are available. Unfortunately, vegetation data are not ye t available. For SWE, the value on the first day of April was averaged over the Middle Rogue sub -basin. For streamflow, monthly sums of total daily runoff were averaged over the Middle Rogue. For each variable except runoff, we generated an annual time ser ies from 1950 to 2099 and computed time period averages for each model and scenario. In Table 4, we p resent future changes from the historical period (1950 -2005 average ) to the 2050s (2040 -2069 average) and the 2080s (2070 -2099 av erage) as a mean and range of the differences computed for each model for a low (RCP4.5) and a high (RCP8.5) emissions scenario. Changes in monthly hydrology of total runoff are presented for the 2080s under both emissions scenarios. Projections for the 20 80s under RCP8.5 for all variables are described in the text.