|Abstract or Summary
- Energy use and greenhouse gas emissions are on the forefront of planning policy in the world today. In the U.S., the transportation sector accounts for 50 percent of greenhouse gas (GHG) emissions, the most of any single sector. State and local levels of government have been very proactive in the mitigation of GHG, with Washington State as one of the leaders. Two state laws passed in 2008 mandate a reduction in GHG and vehicle miles traveled (VMT), a primary metric in measuring traffic. This research focuses on Skagit County, Washington, as we attempt to model and quantify both GHG and VMT as they are affected by land use development.
This research relies heavily on the Envision software platform as it was used a project involving an agent based model of alternative future landscapes. The project provided spatial population and employment data in the context of various future development scenarios such as the compact development of the Ecosystem Scenario where the goal was 90% growth within urban growth areas. On the other extreme was the Development Scenario where up to 40% of the growth was allowed in remote and non-incorporated areas.
A majority of the research of this thesis document is devoted to the development of a traffic model structured around the commute of 17 population centers to job centers within Skagit County. The population centers are defined by traffic area zones, or TAZ, that encompass the entire county. The model is unique in traffic modeling literature for its small number of population and destination centers. The countywide accuracy of the model is exemplary at -0.87% standard error relative to current Highway Performance Monitoring System VMT data, although it should be understood that this result has not been repeated by applying the same methodology to other counties.
Given that buildings energy use is second to the transportation sector as the largest single contributor to GHG emissions, a residential building model was created, in which parcel level population densities characterize high and low density building development. Defined as Apartments and Houses, respectively, within the text, energy use values were assigned to each from climate-specific data of the Residential Energy Consumption Survey national data set. Thus, as population and employment grow in various development scenarios of the Envision-Skagit 2060 project, the GHG generation from buildings and vehicles is calculated for comparison.
The compact development of the Ecosystem Scenario generated the least amount of GHG in both models. Only a 5.9% difference was found in GHGs generated from the building model in the two extreme growth scenarios (Ecosystem and Development). In the traffic model, a 19% difference was found in the VMT of the same two scenarios. Compact, near job center development creates more of an impact on GHG inventory in the transportation sector at approximately twice the GHG impact of compact buildings.
Comparing our results to Washington State law mandated reductions for both VMT and GHGs, we find that neither will be met in the transportation or building sectors. When a applying the Energy Information Association’s most efficient vehicle fuel efficiency scenario where all vehicles are projected to average 59.6 mpg we find that even the best combination of scenarios (Ecosystem land-use and 59.3 mpg vehicle efficiency) results in GHGs that are 1.8 times larger than those permissible by the law as applied to year 2050.
Our results assume the same 'standard of living' as today applied to home appliances, heating/cooling, and vehicle use. The traffic model assumes all commuters continue to use all modes of transit in the same fractions as today: 77.4% use single occupancy vehicles, 12.6% use high occupancy vehicles, 1.0% mass transit, 4.6% walk or bike, and 4.4% work at home in Skagit County. Although increased use of mass transportation was not modeled in this research, it may be the only option given our findings, with regards to the Washington State reduction requirements. In any case, our most important result is a novel approach to traffic modeling that requires only spatial knowedge of population and employment in order to predict traffic and quantify GHG emissions as they change with urban form.