Decar­boniz­ing elec­tric­i­ty pro­duc­tion is cen­tral to reduc­ing green­house gas emis­sions. Exploit­ing inter­mit­tent renew­able ener­gy resources demands pow­er sys­tem plan­ning mod­els with high tem­po­ral and spa­tial res­o­lu­tion. We use a mixed-inte­ger lin­ear pro­gram­ming mod­el – SWITCH – to ana­lyze least-cost gen­er­a­tion, stor­age, and trans­mis­sion capac­i­ty expan­sion for west­ern North Amer­i­ca under var­i­ous pol­i­cy and cost sce­nar­ios. Cur­rent renew­able port­fo­lio stan­dards are shown to be insuf­fi­cient to meet emis­sion reduc­tion tar­gets by 2030 with­out new pol­i­cy. With stronger car­bon pol­i­cy con­sis­tent with a 450 ppm cli­mate sta­bi­liza­tion sce­nario, pow­er sec­tor emis­sions can be reduced to 54% of 1990 lev­els by 2030 using dif­fer­ent port­fo­lios of exist­ing gen­er­a­tion tech­nolo­gies. Under a range of resource cost sce­nar­ios, most coal pow­er plants would be replaced by solar, wind, gas, and/​or nuclear gen­er­a­tion, with inter­mit­tent renew­able sources pro­vid­ing at least 17% and as much as 29% of total pow­er by 2030. The car­bon price to induce these deep car­bon emis­sion reduc­tions is high, but, assum­ing car­bon price rev­enues are rein­vest­ed in the pow­er sec­tor, the cost of pow­er is found to increase by at most 20% rel­a­tive to busi­ness-as-usu­al projections.